Repository: HaonanGuo/Remote-Sensing-ChatGPT Branch: main Commit: 96adff321ce1 Files: 91 Total size: 456.0 KB Directory structure: gitextract_ppmvimf1/ ├── Prefix/ │ └── __init__.py ├── README.md ├── RSChatGPT-shell.py ├── RStask/ │ ├── EdgeDetection/ │ │ ├── Canny.py │ │ ├── __init__.py │ │ └── test.py │ ├── ImageCaptioning/ │ │ ├── __init__.py │ │ ├── blip.py │ │ └── test.py │ ├── InstanceSegmentation/ │ │ ├── SwinUpper.py │ │ ├── __init__.py │ │ ├── model.py │ │ ├── swin.py │ │ ├── test.py │ │ └── uper.py │ ├── LanduseSegmentation/ │ │ ├── __init__.py │ │ ├── seg_hrnet.py │ │ └── test.py │ ├── ObjectCounting/ │ │ ├── Yolocounting.py │ │ ├── __init__.py │ │ └── test.py │ ├── ObjectDetection/ │ │ ├── YOLOv5.py │ │ ├── __init__.py │ │ ├── models/ │ │ │ ├── __init__.py │ │ │ ├── common.py │ │ │ ├── experimental.py │ │ │ ├── hub/ │ │ │ │ ├── anchors.yaml │ │ │ │ ├── yolov3-spp.yaml │ │ │ │ ├── yolov3-tiny.yaml │ │ │ │ ├── yolov3.yaml │ │ │ │ ├── yolov5-bifpn.yaml │ │ │ │ ├── yolov5-fpn.yaml │ │ │ │ ├── yolov5-p2.yaml │ │ │ │ ├── yolov5-p34.yaml │ │ │ │ ├── yolov5-p6.yaml │ │ │ │ ├── yolov5-p7.yaml │ │ │ │ ├── yolov5-panet.yaml │ │ │ │ ├── yolov5l6.yaml │ │ │ │ ├── yolov5m6.yaml │ │ │ │ ├── yolov5n6.yaml │ │ │ │ ├── yolov5s-ghost.yaml │ │ │ │ ├── yolov5s-transformer.yaml │ │ │ │ ├── yolov5s6.yaml │ │ │ │ └── yolov5x6.yaml │ │ │ ├── tf.py │ │ │ ├── yolo.py │ │ │ ├── yolov5l.yaml │ │ │ ├── yolov5m.yaml │ │ │ ├── yolov5n.yaml │ │ │ ├── yolov5s.yaml │ │ │ └── yolov5x.yaml │ │ ├── test.py │ │ └── utils/ │ │ ├── __init__.py │ │ ├── activations.py │ │ ├── augmentations.py │ │ ├── autoanchor.py │ │ ├── autobatch.py │ │ ├── aws/ │ │ │ ├── __init__.py │ │ │ ├── mime.sh │ │ │ ├── resume.py │ │ │ └── userdata.sh │ │ ├── benchmarks.py │ │ ├── callbacks.py │ │ ├── datasets.py │ │ ├── docker/ │ │ │ ├── Dockerfile │ │ │ └── Dockerfile-cpu │ │ ├── downloads.py │ │ ├── flask_rest_api/ │ │ │ ├── README.md │ │ │ ├── example_request.py │ │ │ └── restapi.py │ │ ├── general.py │ │ ├── google_app_engine/ │ │ │ ├── Dockerfile │ │ │ ├── additional_requirements.txt │ │ │ └── app.yaml │ │ ├── loggers/ │ │ │ ├── __init__.py │ │ │ └── wandb/ │ │ │ ├── README.md │ │ │ ├── __init__.py │ │ │ ├── log_dataset.py │ │ │ ├── sweep.py │ │ │ ├── sweep.yaml │ │ │ └── wandb_utils.py │ │ ├── loss.py │ │ ├── metrics.py │ │ ├── plots.py │ │ └── torch_utils.py │ ├── SceneClassification/ │ │ ├── ResNetScene.py │ │ ├── __init__.py │ │ └── test.py │ └── __init__.py ├── checkpoints/ │ └── __init__.py └── requirements.txt ================================================ FILE CONTENTS ================================================ ================================================ FILE: Prefix/__init__.py ================================================ RS_CHATGPT_PREFIX = """Remote Sensing ChatGPT is designed to assist with a wide range of remote sensing image related tasks, from answering simple questions to providing in-depth explanations and discussions on a wide range of remote sensing applications. Remote Sensing ChatGPT is able to generate human-like text based on the input it receives, allowing it to engage in natural-sounding conversations and provide responses that are coherent and relevant to the topic at hand. Remote Sensing ChatGPT can process and understand large amounts of remote sensing images, knowledge, and text. As a expertized language model, Remote Sensing ChatGPT can not directly read remote sensing images, but it has a list of tools to finish different remote sensing tasks. Each input remote sensing image will have a file name formed as "image/xxx.png", and Remote Sensing ChatGPT can invoke different tools to indirectly understand the remote sensing image. When talking about images, Remote Sensing ChatGPT is very strict to the file name and will never fabricate nonexistent files. When using tools to generate new image files, Remote Sesning ChatGPT is also known that the image may not be the same as the user's demand, and will use other visual question answering tools or description tools to observe the real image. Remote Sensing ChatGPT is able to use tools in a sequence, and is loyal to the tool observation outputs rather than faking the image content and image file name. It will remember to provide the file name from the last tool observation, if a new image is generated. Human may provide new remote sensing images to Remote Sensing ChatGPT with a description. The description helps Remote Sensing ChatGPT to understand this image, but Remote Sensing ChatGPT should use tools to finish following tasks, rather than directly imagine from the description. Overall, Remote Sensing ChatGPT is a powerful visual dialogue assistant tool that can help with a wide range of remote sensing tasks and provide valuable insights and information on a wide range of remote sensing applicatinos. TOOLS: ------ Remote Sensing ChatGPT has access to the following tools:""" RS_CHATGPT_FORMAT_INSTRUCTIONS = """ When you have a response to say to the Human, or if you do not need to use a tool, you MUST use the format: ``` Thought: Do I need to use a tool? No {ai_prefix}: [your response here] ``` To use a tool, you MUST use the following format: ``` Thought: Do I need to use a tool? Yes Action: the action to take, should be one of [{tool_names}] Action Input: the input to the action Observation: the result of the action ``` """ RS_CHATGPT_SUFFIX = """You are very strict to the filename correctness and will never fake a file name if it does not exist. You will remember to provide the image file name loyally if it's provided in the last tool observation. Begin! Previous conversation history: {chat_history} New input: {input} Since Remote Sensing ChatGPT is a text language model, Remote Sensing ChatGPT must use tools to observe remote sensing images rather than imagination. The thoughts and observations are only visible for Remote Sensing ChatGPT, Remote Sensing ChatGPT should remember to repeat important information in the final response for Human. Thought: Do I need to use a tool? {agent_scratchpad} Let's think step by step. """ ================================================ FILE: README.md ================================================ # RS-ChatGPT: Solving Remote Sensing Tasks with ChatGPT and Visual Models Introduction ---- Remote Sensing ChatGPT(RS-ChatGPT) is an open source tool for solving remote sensing tasks with ChatGPT in an interactive way.ChatGPT acts as an expert to response to users' linguistic resquests based on the input remote sensing image. It supports various interpretation tasks that are trained on remote sensing datasets. To help ChatGPT better understand remote sensing knowledge, remote sensing image captioning is set as cue to help ChatGPT understand remote sensing image.We will keep updating RS-CHatGPT~ Please comments with Issues or send me a email if you have any suggestions! Thanks in advance! Generally, RS-ChatGPT includes four steps in implementation: * Prompt Template Generation * Task Planning * Task Execution * Response Genration Our article is now available in [arXiv!](https://arxiv.org/abs/2401.09083) See you in IGARSS204, Greece!
My personal website is also avalable:[www.poleguo98.top](https://www.poleguo98.top) Updates ---- * TODO1: An online website is comming! Let's use RS-ChatGPT Online!
* TODO2: What about asking RS-CHatGPT to choose different models within the given task?
* TODO3: Put all the models available in Huggingface
2024.01.30:
* Remote Sensing GPT is updated, with bugs fixed, more readable, and more scalable~
2023.09.19:
* Remote Sensing GPT now supports GPT-4 and multi-round chating! An interactive interface is now available(see interface.py)~
2023.08.23:
* Initial release The code ---- ### Requirements Please Refer to [requirements.txt](https://github.com/HaonanGuo/Remote-Sensing-ChatGPT/blob/main/requirements.txt) ### Usage ->Clone the repository:git clone https://github.com/HaonanGuo/Remote-Sensing-ChatGPT
->Download the below models and place them in the checkpoints folder
->Run [RSChatGPT-shell.py](https://github.com/HaonanGuo/Remote-Sensing-ChatGPT/blob/main/RSChatGPT-shell.py) ### Supported Function | Function | Description | Method | Pretrain Dataset | Model Weights | | :--------: | :--------: | :--------: | :--------: | :--------: | | Image Captioning | Describe the remote sensing image | [BLIP](https://icml.cc/virtual/2022/spotlight/16016) | [BLIP Dataset](https://icml.cc/virtual/2022/spotlight/16016)| [weight(github)](https://github.com/salesforce/BLIP) | | Scene Classification | Classify the type of scene | [ResNet](https://arxiv.org/abs/1512.03385) | [AID Dataset](http://www.captain-whu.com/project/AID/)|[weight(Google)](https://drive.google.com/file/d/1f-WES6fTGGa5W9BcDPMVhGk3Foc4p9Or/view?usp=drive_link) [weight(Baidu)](https://pan.baidu.com/s/1yNgUQKieZBEJZ0axzN4tiw?pwd=RSGP) | | Object Detection | Detect RS object from image | [YOLO v5](https://zenodo.org/badge/latestdoi/264818686) | [DOTA](http://captain.whu.edu.cn/DOTAweb)| [weight(Google)](https://drive.google.com/file/d/1Hb7XA6gZxNam8y8nxs2p6EqJ-XaG1o5Y/view?usp=drive_link) [weight(Baidu)](https://pan.baidu.com/s/1XTG-MLxx5_D0OO6M80OP1A?pwd=RSGP) | | Instance Segmentation | Extract Instance Mask of certain object | [SwinTransformer+UperNet](https://github.com/open-mmlab/mmsegmentation) | [iSAID](https://captain-whu.github.io/iSAID/index)| [weight(Google)](https://drive.google.com/file/d/165jeD0oi6fSpvWrpgfVBbzUOsyHN0xEq/view?usp=drive_link) [weight(Baidu)](https://pan.baidu.com/s/1Tv6BCt68L2deY_wMVZizgg?pwd=RSGP)| | Landuse Classification | Extract Pixel-wise Landuse Classification | [HRNet](https://github.com/HRNet) | [LoveDA](https://github.com/Junjue-Wang/LoveDA)| [weight(Google)](https://drive.google.com/file/d/1fRyEpb7344S4Y5F2Q4EBO3fXVT4kXaft/view?usp=drive_link) [weight(Baidu)](https://pan.baidu.com/s/1m6yOXbT6cKGqJ64z86u7fQ?pwd=RSGP) | | Object Counting | Count the number of certain object in an image | [YOLO v5](https://zenodo.org/badge/latestdoi/264818686) | [DOTA](http://captain.whu.edu.cn/DOTAweb)| Same as Object Detection | | Edge Detection | Extract edge of remote sensing image | Canny |None| None | More funtions to be updated~ ### Citation Please cite the repo if you use the data or code in this repo. ``` @article{RS ChatGPT, title = {Remote Sensing ChatGPT: Solving Remote Sensing Tasks with ChatGPT and Visual Models}, shorttitle = {Remote Sensing ChatGPT}, doi = {10.48550/ARXIV.2401.09083}, author = {Guo, Haonan and Su, Xin and Wu, Chen and Du, Bo and Zhang, Liangpei and Li, Deren}, year = {2024}, } ``` ## Acknowledgments - [Visual ChatGPT](https://github.com/microsoft/TaskMatrix) - [YOLOv5](https://github.com/hukaixuan19970627/yolov5_obb) - [BLIP](https://github.com/salesforce/BLIP) Help ---- Remote Sensing ChatGPT is an open source project that welcome any contribution and feedback. Please contact us with: haonan.guo@whu.edu.cn ================================================ FILE: RSChatGPT-shell.py ================================================ import os import re import uuid from skimage import io import argparse import inspect from langchain.chat_models import ChatOpenAI from langchain.agents.initialize import initialize_agent from langchain.agents.tools import Tool from langchain.chains.conversation.memory import ConversationBufferMemory import numpy as np from Prefix import RS_CHATGPT_PREFIX, RS_CHATGPT_FORMAT_INSTRUCTIONS, RS_CHATGPT_SUFFIX from RStask import ImageEdgeFunction,CaptionFunction,LanduseFunction,DetectionFunction,CountingFuncnction,SceneFunction,InstanceFunction os.makedirs('image', exist_ok=True) def prompts(name, description): def decorator(func): func.name = name func.description = description return func return decorator def get_new_image_name(org_img_name, func_name="update"): head_tail = os.path.split(org_img_name) head = head_tail[0] tail = head_tail[1] name_split = tail.split('.')[0].split('_') this_new_uuid = str(uuid.uuid4())[:4] recent_prev_file_name = name_split[0] new_file_name = f'{this_new_uuid}_{func_name}_{recent_prev_file_name}.png'.replace('__','_') return os.path.join(head, new_file_name) class EdgeDetection: def __init__(self, device): print("Initializing Edge Detection Function....") self.func = ImageEdgeFunction() @prompts(name="Edge Detection On Image", description="useful when you want to detect the edge of the remote sensing image. " "like: detect the edges of this image, or canny detection on image, " "or perform edge detection on this image, or detect the edge of this image. " "The input to this tool should be a string, representing the image_path") def inference(self, inputs): updated_image_path=get_new_image_name(inputs, func_name="edge") self.func.inference(inputs,updated_image_path) return updated_image_path class ObjectCounting: def __init__(self, device): self.func=CountingFuncnction(device) @prompts(name="Count object", description="useful when you want to count the number of the object in the image. " "like: how many planes are there in the image? or count the number of bridges" "The input to this tool should be a comma separated string of two, " "representing the image_path, the text description of the object to be counted") def inference(self, inputs): image_path, det_prompt = inputs.split(",") log_text=self.func.inference(image_path,det_prompt) return log_text class InstanceSegmentation: def __init__(self, device): print("Initializing InstanceSegmentation") self.func=InstanceFunction(device) @prompts(name="Instance Segmentation for Remote Sensing Image", description="useful when you want to apply man-made instance segmentation for the image. The expected input category include plane, ship, storage tank, baseball diamond, tennis court, basketball court, ground track field, harbor, bridge, vehicle, helicopter, roundabout, soccer ball field, and swimming pool." "like: extract plane from this image, " "or predict the ship in this image, or extract tennis court from this image, segment harbor from this image, Extract the vehicle in the image. " "The input to this tool should be a comma separated string of two, " "representing the image_path, the text of the category,selected from plane, or ship, or storage tank, or baseball diamond, or tennis court, or basketball court, or ground track field, or harbor, or bridge, or vehicle, or helicopter, or roundabout, or soccer ball field, or swimming pool. ") def inference(self, inputs): image_path, det_prompt = inputs.split(",") updated_image_path = get_new_image_name(image_path, func_name="instance_" + det_prompt) text=self.func.inference(image_path, det_prompt,updated_image_path) return text class SceneClassification: def __init__(self, device): print("Initializing SceneClassification") self.func=SceneFunction(device) @prompts(name="Scene Classification for Remote Sensing Image", description="useful when you want to know the type of scene or function for the image. " "like: what is the category of this image?, " "or classify the scene of this image, or predict the scene category of this image, or what is the function of this image. " "The input to this tool should be a string, representing the image_path. ") def inference(self, inputs): output_txt=self.func.inference(inputs) return output_txt class LandUseSegmentation: def __init__(self, device): print("Initializing LandUseSegmentation") self.func=LanduseFunction(device) @prompts(name="Land Use Segmentation for Remote Sensing Image", description="useful when you want to apply land use gegmentation for the image. The expected input category include Building, Road, Water, Barren, Forest, Farmland, Landuse." "like: generate landuse map from this image, " "or predict the landuse on this image, or extract building from this image, segment roads from this image, Extract the water bodies in the image. " "The input to this tool should be a comma separated string of two, " "representing the image_path, the text of the category,selected from Lnad Use, or Building, or Road, or Water, or Barren, or Forest, or Farmland, or Landuse.") def inference(self, inputs): image_path, det_prompt = inputs.split(",") updated_image_path = get_new_image_name(image_path, func_name="landuse") text=self.func.inference(image_path, det_prompt,updated_image_path) return text class ObjectDetection: def __init__(self, device): self.func=DetectionFunction(device) @prompts(name="Detect the given object", description="useful when you only want to detect the bounding box of the certain objects in the picture according to the given text." "like: detect the plane, or can you locate an object for me." "The input to this tool should be a comma separated string of two, " "representing the image_path, the text description of the object to be found") def inference(self, inputs): image_path, det_prompt = inputs.split(",") updated_image_path = get_new_image_name(image_path, func_name="detection_" + det_prompt.replace(' ', '_')) log_text=self.func.inference(image_path, det_prompt,updated_image_path) return log_text class ImageCaptioning: def __init__(self, device): print(f"Initializing ImageCaptioning to {device}") self.device = device self.func=CaptionFunction(device) @prompts(name="Get Photo Description", description="useful when you want to know what is inside the photo. receives image_path as input. " "The input to this tool should be a string, representing the image_path. ") def inference(self, image_path): captions = self.func.inference(image_path) print(f"\nProcessed ImageCaptioning, Input Image: {image_path}, Output Text: {captions}") return captions class RSChatGPT: def __init__(self, gpt_name,load_dict,openai_key,proxy_url): print(f"Initializing RSChatGPT, load_dict={load_dict}") if 'ImageCaptioning' not in load_dict: raise ValueError("You have to load ImageCaptioning as a basic function for RSChatGPT") self.models = {} # Load Basic Foundation Models for class_name, device in load_dict.items(): self.models[class_name] = globals()[class_name](device=device) # Load Template Foundation Models for class_name, module in globals().items(): if getattr(module, 'template_model', False): template_required_names = {k for k in inspect.signature(module.__init__).parameters.keys() if k != 'self'} loaded_names = set([type(e).__name__ for e in self.models.values()]) if template_required_names.issubset(loaded_names): self.models[class_name] = globals()[class_name]( **{name: self.models[name] for name in template_required_names}) print(f"All the Available Functions: {self.models}") self.tools = [] for instance in self.models.values(): for e in dir(instance): if e.startswith('inference'): func = getattr(instance, e) self.tools.append(Tool(name=func.name, description=func.description, func=func)) self.llm = ChatOpenAI(api_key=openai_key, base_url=proxy_url, model_name=gpt_name,temperature=0) self.memory = ConversationBufferMemory(memory_key="chat_history", output_key='output') def initialize(self): self.memory.clear() #clear previous history PREFIX, FORMAT_INSTRUCTIONS, SUFFIX = RS_CHATGPT_PREFIX, RS_CHATGPT_FORMAT_INSTRUCTIONS, RS_CHATGPT_SUFFIX self.agent = initialize_agent( self.tools, self.llm, agent="conversational-react-description", verbose=True, memory=self.memory, return_intermediate_steps=True,stop=["\nObservation:", "\n\tObservation:"], agent_kwargs={'prefix': PREFIX, 'format_instructions': FORMAT_INSTRUCTIONS,'suffix': SUFFIX}, ) def run_text(self, text, state): res = self.agent({"input": text.strip()}) res['output'] = res['output'].replace("\\", "/") response = re.sub('(image/[-\w]*.png)', lambda m: f'![](file={m.group(0)})*{m.group(0)}*', res['output']) state = state + [(text, response)] print(f"\nProcessed run_text, Input text: {text}\nCurrent state: {state}\n" f"Current Memory: {self.agent.memory.buffer}") return state def run_image(self, image_dir, state, txt=None): image_filename = os.path.join('image', f"{str(uuid.uuid4())[:8]}.png") img = io.imread(image_dir) # width, height = img.shape[1],img.shape[0] # ratio = min(640 / width, 640 / height) # if ratio<1: # width_new, height_new = (round(width * ratio), round(height * ratio)) # else: # width_new, height_new =width,height # width_new = int(np.round(width_new / 64.0)) * 64 # height_new = int(np.round(height_new / 64.0)) * 64 # # if width_new!=width or height_new!=height: # img = cv2.resize(img,(width_new, height_new)) # print(f"======>Auto Resizing Image from {height,width} to {height_new,width_new}...") # else: # print(f"======>Auto Renaming Image...") io.imsave(image_filename, img.astype(np.uint8)) description = self.models['ImageCaptioning'].inference(image_filename) Human_prompt = f' Provide a remote sensing image named {image_filename}. The description is: {description}. This information helps you to understand this image, but you should use tools to finish following tasks, rather than directly imagine from my description. If you understand, say \"Received\".' AI_prompt = "Received." self.memory.chat_memory.add_user_message(Human_prompt) self.memory.chat_memory.add_ai_message(AI_prompt) state = state + [(f"![](file={image_filename})*{image_filename}*", AI_prompt)] print(f"\nProcessed run_image, Input image: {image_filename}\nCurrent state: {state}\n" f"Current Memory: {self.agent.memory.buffer}") state=self.run_text(f'{txt} {image_filename} ', state) return state if __name__ == '__main__': parser = argparse.ArgumentParser() parser.add_argument('--openai_key', type=str,required=True) parser.add_argument('--image_dir', type=str,required=True) parser.add_argument('--gpt_name', type=str, default="gpt-3.5-turbo",choices=['gpt-3.5-turbo-1106','gpt-3.5-turbo','gpt-4','gpt-4-0125-preview','gpt-4-turbo-preview','gpt-4-1106-preview']) parser.add_argument('--proxy_url', type=str, default=None) parser.add_argument('--load', type=str,help='Image Captioning is basic models that is required. You can select from [ImageCaptioning,ObjectDetection,LandUseSegmentation,InstanceSegmentation,ObjectCounting,SceneClassification,EdgeDetection]', default="ImageCaptioning_cuda:0,SceneClassification_cuda:0,ObjectDetection_cuda:0,LandUseSegmentation_cuda:0,InstanceSegmentation_cuda:0,ObjectCounting_cuda:0,EdgeDetection_cpu") args = parser.parse_args() state = [] load_dict = {e.split('_')[0].strip(): e.split('_')[1].strip() for e in args.load.split(',')} bot = RSChatGPT(gpt_name=args.gpt_name,load_dict=load_dict,openai_key=args.openai_key,proxy_url=args.proxy_url) bot.initialize() print('RSChatGPT initialization done, you can now chat with RSChatGPT~') bot.initialize() txt='Count the number of plane in the image.' state=bot.run_image(args.image_dir, [], txt) while 1: txt = input('You can now input your question.(e.g. Extract buildings from the image)\n') state = bot.run_image(args.image_dir, state, txt) ================================================ FILE: RStask/EdgeDetection/Canny.py ================================================ from PIL import Image import cv2 import numpy as np class Image2Canny: def __init__(self): print("Initializing Image2Canny") self.low_threshold = 100 self.high_threshold = 200 def inference(self, inputs,new_image_name): image = Image.open(inputs) image = np.array(image) canny = cv2.Canny(image, self.low_threshold, self.high_threshold) canny = canny[:, :, None] canny = np.concatenate([canny, canny, canny], axis=2) canny = Image.fromarray(canny) updated_image_path = new_image_name canny.save(updated_image_path) print(f"\nProcessed Image2Canny, Input Image: {inputs}, Output Text: {updated_image_path}") return None ================================================ FILE: RStask/EdgeDetection/__init__.py ================================================ ================================================ FILE: RStask/EdgeDetection/test.py ================================================ from RStask.EdgeDetection.Canny import Image2Canny model=Image2Canny() model.inference('/data/haonan.guo/RSChatGPT/test.tif','/data/haonan.guo/RSChatGPT/output.png') ================================================ FILE: RStask/ImageCaptioning/__init__.py ================================================ from RStask.ImageCaptioning.blip import BLIP as CaptionFunction ================================================ FILE: RStask/ImageCaptioning/blip.py ================================================ import torch from PIL import Image from transformers import BlipProcessor, BlipForConditionalGeneration class BLIP: def __init__(self, device): self.device = device self.torch_dtype = torch.float16 if 'cuda' in device else torch.float32 self.processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base") self.model = BlipForConditionalGeneration.from_pretrained( "Salesforce/blip-image-captioning-base", torch_dtype=self.torch_dtype).to(self.device) def inference(self, image_path): inputs = self.processor(Image.open(image_path), return_tensors="pt").to(self.device, self.torch_dtype) out = self.model.generate(**inputs) captions = 'A satellite image of ' + self.processor.decode(out[0], skip_special_tokens=True) print(f"\nProcessed ImageCaptioning, Input Image: {image_path}, Output Text: {captions}") return captions ================================================ FILE: RStask/ImageCaptioning/test.py ================================================ from RStask import LanduseFunction model=LanduseFunction('cuda:0') model.inference('/data/haonan.guo/LoveDA/Train/Urban/images_png/1367.png','building','/data/haonan.guo/RSChatGPT/output.png') ================================================ FILE: RStask/InstanceSegmentation/SwinUpper.py ================================================ from RStask.InstanceSegmentation.model import SwinUPer import torch from skimage import io from PIL import Image import numpy as np class SwinInstance: def __init__(self, device): print("Initializing InstanceSegmentation") self.model = SwinUPer() self.device = device try: trained = torch.load('./checkpoints/last_swint_upernet_finetune.pth') except: trained = torch.load('../../checkpoints/last_swint_upernet_finetune.pth') self.model.load_state_dict(trained["state_dict"]) self.model = self.model.to(device) self.model.eval() self.mean, self.std = torch.tensor([123.675, 116.28, 103.53]).reshape((1, 3, 1, 1)), torch.tensor( [58.395, 57.12, 57.375]).reshape((1, 3, 1, 1)) self.all_dict = {'plane': 1, 'ship': 2, 'storage tank': 3, 'baseball diamond': 4, 'tennis court': 5, 'basketball court': 6, 'ground track field': 7, 'harbor': 8, 'bridge': 9, 'large vehicle': 10, 'small vehicle': 11, 'helicopter': 12, 'roundabout': 13, 'soccer ball field': 14, 'swimming pool': 15} def inference(self, image_path, det_prompt ,updated_image_path): image = torch.from_numpy(io.imread(image_path)) image = (image.permute(2, 0, 1).unsqueeze(0) - self.mean) / self.std with torch.no_grad(): pred = self.model(image.to(self.device)) pred = pred.argmax(1).cpu().squeeze().int().numpy() if det_prompt.strip().lower() in [i.strip().lower() for i in self.all_dict.keys()]: idx=[i.replace(' ', '_').lower() for i in self.all_dict.keys()].index(det_prompt.strip().lower())+1 pred=(pred==idx)*255 pred = Image.fromarray(np.stack([pred, pred, pred], -1).astype(np.uint8)) pred.save(updated_image_path) print(f"\nProcessed Instance Segmentation, Input Image: {image_path + ',' + det_prompt}, Output SegMap: {updated_image_path}") return updated_image_path else: print(f"\nCategory: { det_prompt} is not supported. Please use other tools.") return f"Category {det_prompt} is not supported. Please use other tools." ================================================ FILE: RStask/InstanceSegmentation/__init__.py ================================================ ================================================ FILE: RStask/InstanceSegmentation/model.py ================================================ import torch import torch.nn as nn from RStask.InstanceSegmentation.swin import swin from RStask.InstanceSegmentation.uper import UPerHead Activation=torch.nn.ReLU class SegmentationHead(nn.Sequential): def __init__(self, in_channels, out_channels, kernel_size=3, activation=None, upsampling=1): conv2d = nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, padding=kernel_size // 2) upsampling = nn.UpsamplingBilinear2d(scale_factor=upsampling) if upsampling > 1 else nn.Identity() activation = Activation(activation) super().__init__(conv2d, upsampling, activation) class ClassificationHead(nn.Sequential): def __init__(self, in_channels, classes, pooling="avg", dropout=0.2, activation=None): if pooling not in ("max", "avg"): raise ValueError("Pooling should be one of ('max', 'avg'), got {}.".format(pooling)) pool = nn.AdaptiveAvgPool2d(1) if pooling == "avg" else nn.AdaptiveMaxPool2d(1) flatten = nn.Flatten() dropout = nn.Dropout(p=dropout, inplace=True) if dropout else nn.Identity() linear = nn.Linear(in_channels, classes, bias=True) activation = Activation(activation) super().__init__(pool, flatten, dropout, linear, activation) def initialize_decoder(module): for m in module.modules(): if isinstance(m, nn.Conv2d): nn.init.kaiming_uniform_(m.weight, mode="fan_in", nonlinearity="relu") if m.bias is not None: nn.init.constant_(m.bias, 0) elif isinstance(m, nn.BatchNorm2d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) elif isinstance(m, nn.Linear): nn.init.xavier_uniform_(m.weight) if m.bias is not None: nn.init.constant_(m.bias, 0) def initialize_head(module): for m in module.modules(): if isinstance(m, (nn.Linear, nn.Conv2d)): nn.init.xavier_uniform_(m.weight) if m.bias is not None: nn.init.constant_(m.bias, 0) class SwinUPer(torch.nn.Module): def __init__(self, classes: int = 16): super(SwinUPer, self).__init__() # encoder self.encoder = swin(embed_dim=96,depths=[2, 2, 6, 2],num_heads=[3, 6, 12, 24], window_size=7,ape=False,drop_path_rate=0.3,patch_norm=True) # decoder self.decoder = UPerHead( in_channels = self.encoder.out_channels[1:], channels = self.encoder.out_channels[2], in_index = (0, 1, 2, 3),dropout_ratio = 0.1, norm_cfg = dict(type='SyncBN', requires_grad=True) ) self.semseghead = nn.Sequential( nn.Dropout2d(0.1), nn.Conv2d(self.encoder.out_channels[2], classes, kernel_size=1) ) self.initialize() def initialize(self): initialize_decoder(self.decoder) initialize_head(self.semseghead) def forward(self, x): features = self.encoder(x) output = self.decoder(*features) output = self.semseghead(output) return output ================================================ FILE: RStask/InstanceSegmentation/swin.py ================================================ # -------------------------------------------------------- # Swin Transformer # Copyright (c) 2021 Microsoft # Licensed under The MIT License [see LICENSE for details] # Written by Ze Liu, Yutong Lin, Yixuan Wei # -------------------------------------------------------- import warnings from collections import OrderedDict import torch import torch.nn as nn import torch.nn.functional as F import torch.utils.checkpoint as checkpoint import numpy as np from timm.models.layers import DropPath, to_2tuple, trunc_normal_ #from custom import load_checkpoint #from mmseg.utils import get_root_logger #from ..builder import BACKBONES #from mmcv.cnn.utils.weight_init import (constant_init, trunc_normal_, # trunc_normal_init) #from mmcv.runner import BaseModule, ModuleList, _load_checkpoint class Mlp(nn.Module): """ Multilayer perceptron.""" def __init__(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.): super().__init__() out_features = out_features or in_features hidden_features = hidden_features or in_features self.fc1 = nn.Linear(in_features, hidden_features) self.act = act_layer() self.fc2 = nn.Linear(hidden_features, out_features) self.drop = nn.Dropout(drop) def forward(self, x): x = self.fc1(x) x = self.act(x) x = self.drop(x) x = self.fc2(x) x = self.drop(x) return x def window_partition(x, window_size): """ Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C) """ B, H, W, C = x.shape x = x.view(B, H // window_size, window_size, W // window_size, window_size, C) windows = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, window_size, window_size, C) return windows def window_reverse(windows, window_size, H, W): """ Args: windows: (num_windows*B, window_size, window_size, C) window_size (int): Window size H (int): Height of image W (int): Width of image Returns: x: (B, H, W, C) """ B = int(windows.shape[0] / (H * W / window_size / window_size)) x = windows.view(B, H // window_size, W // window_size, window_size, window_size, -1) x = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(B, H, W, -1) return x class WindowAttention(nn.Module): """ Window based multi-head self attention (W-MSA) module with relative position bias. It supports both of shifted and non-shifted window. Args: dim (int): Number of input channels. window_size (tuple[int]): The height and width of the window. num_heads (int): Number of attention heads. qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set attn_drop (float, optional): Dropout ratio of attention weight. Default: 0.0 proj_drop (float, optional): Dropout ratio of output. Default: 0.0 """ def __init__(self, dim, window_size, num_heads, qkv_bias=True, qk_scale=None, attn_drop=0., proj_drop=0.): super().__init__() self.dim = dim self.window_size = window_size # Wh, Ww self.num_heads = num_heads head_dim = dim // num_heads self.scale = qk_scale or head_dim ** -0.5 # define a parameter table of relative position bias self.relative_position_bias_table = nn.Parameter( torch.zeros((2 * window_size[0] - 1) * (2 * window_size[1] - 1), num_heads)) # 2*Wh-1 * 2*Ww-1, nH # get pair-wise relative position index for each token inside the window coords_h = torch.arange(self.window_size[0]) coords_w = torch.arange(self.window_size[1]) coords = torch.stack(torch.meshgrid([coords_h, coords_w])) # 2, Wh, Ww coords_flatten = torch.flatten(coords, 1) # 2, Wh*Ww relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :] # 2, Wh*Ww, Wh*Ww relative_coords = relative_coords.permute(1, 2, 0).contiguous() # Wh*Ww, Wh*Ww, 2 relative_coords[:, :, 0] += self.window_size[0] - 1 # shift to start from 0 relative_coords[:, :, 1] += self.window_size[1] - 1 relative_coords[:, :, 0] *= 2 * self.window_size[1] - 1 relative_position_index = relative_coords.sum(-1) # Wh*Ww, Wh*Ww self.register_buffer("relative_position_index", relative_position_index) self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias) self.attn_drop = nn.Dropout(attn_drop) self.proj = nn.Linear(dim, dim) self.proj_drop = nn.Dropout(proj_drop) trunc_normal_(self.relative_position_bias_table, std=.02) self.softmax = nn.Softmax(dim=-1) def forward(self, x, mask=None): """ Forward function. Args: x: input features with shape of (num_windows*B, N, C) mask: (0/-inf) mask with shape of (num_windows, Wh*Ww, Wh*Ww) or None """ B_, N, C = x.shape qkv = self.qkv(x).reshape(B_, N, 3, self.num_heads, C // self.num_heads).permute(2, 0, 3, 1, 4) q, k, v = qkv[0], qkv[1], qkv[2] # make torchscript happy (cannot use tensor as tuple) q = q * self.scale attn = (q @ k.transpose(-2, -1)) relative_position_bias = self.relative_position_bias_table[self.relative_position_index.view(-1)].view( self.window_size[0] * self.window_size[1], self.window_size[0] * self.window_size[1], -1) # Wh*Ww,Wh*Ww,nH relative_position_bias = relative_position_bias.permute(2, 0, 1).contiguous() # nH, Wh*Ww, Wh*Ww attn = attn + relative_position_bias.unsqueeze(0) if mask is not None: nW = mask.shape[0] attn = attn.view(B_ // nW, nW, self.num_heads, N, N) + mask.unsqueeze(1).unsqueeze(0) attn = attn.view(-1, self.num_heads, N, N) attn = self.softmax(attn) else: attn = self.softmax(attn) attn = self.attn_drop(attn) x = (attn @ v).transpose(1, 2).reshape(B_, N, C) x = self.proj(x) x = self.proj_drop(x) return x class SwinTransformerBlock(nn.Module): """ Swin Transformer Block. Args: dim (int): Number of input channels. num_heads (int): Number of attention heads. window_size (int): Window size. shift_size (int): Shift size for SW-MSA. mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set. drop (float, optional): Dropout rate. Default: 0.0 attn_drop (float, optional): Attention dropout rate. Default: 0.0 drop_path (float, optional): Stochastic depth rate. Default: 0.0 act_layer (nn.Module, optional): Activation layer. Default: nn.GELU norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm """ def __init__(self, dim, num_heads, window_size=7, shift_size=0, mlp_ratio=4., qkv_bias=True, qk_scale=None, drop=0., attn_drop=0., drop_path=0., act_layer=nn.GELU, norm_layer=nn.LayerNorm): super().__init__() self.dim = dim self.num_heads = num_heads self.window_size = window_size self.shift_size = shift_size self.mlp_ratio = mlp_ratio assert 0 <= self.shift_size < self.window_size, "shift_size must in 0-window_size" self.norm1 = norm_layer(dim) self.attn = WindowAttention( dim, window_size=to_2tuple(self.window_size), num_heads=num_heads, qkv_bias=qkv_bias, qk_scale=qk_scale, attn_drop=attn_drop, proj_drop=drop) self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity() self.norm2 = norm_layer(dim) mlp_hidden_dim = int(dim * mlp_ratio) self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=act_layer, drop=drop) self.H = None self.W = None def forward(self, x, mask_matrix): """ Forward function. Args: x: Input feature, tensor size (B, H*W, C). H, W: Spatial resolution of the input feature. mask_matrix: Attention mask for cyclic shift. """ B, L, C = x.shape H, W = self.H, self.W assert L == H * W, "input feature has wrong size" shortcut = x x = self.norm1(x) x = x.view(B, H, W, C) # pad feature maps to multiples of window size pad_l = pad_t = 0 pad_r = (self.window_size - W % self.window_size) % self.window_size pad_b = (self.window_size - H % self.window_size) % self.window_size x = F.pad(x, (0, 0, pad_l, pad_r, pad_t, pad_b)) _, Hp, Wp, _ = x.shape # cyclic shift if self.shift_size > 0: shifted_x = torch.roll(x, shifts=(-self.shift_size, -self.shift_size), dims=(1, 2)) attn_mask = mask_matrix else: shifted_x = x attn_mask = None # partition windows x_windows = window_partition(shifted_x, self.window_size) # nW*B, window_size, window_size, C x_windows = x_windows.view(-1, self.window_size * self.window_size, C) # nW*B, window_size*window_size, C # W-MSA/SW-MSA attn_windows = self.attn(x_windows, mask=attn_mask) # nW*B, window_size*window_size, C # merge windows attn_windows = attn_windows.view(-1, self.window_size, self.window_size, C) shifted_x = window_reverse(attn_windows, self.window_size, Hp, Wp) # B H' W' C # reverse cyclic shift if self.shift_size > 0: x = torch.roll(shifted_x, shifts=(self.shift_size, self.shift_size), dims=(1, 2)) else: x = shifted_x if pad_r > 0 or pad_b > 0: x = x[:, :H, :W, :].contiguous() x = x.view(B, H * W, C) # FFN x = shortcut + self.drop_path(x) x = x + self.drop_path(self.mlp(self.norm2(x))) return x class PatchMerging(nn.Module): """ Patch Merging Layer Args: dim (int): Number of input channels. norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm """ def __init__(self, dim, norm_layer=nn.LayerNorm): super().__init__() self.dim = dim self.reduction = nn.Linear(4 * dim, 2 * dim, bias=False) self.norm = norm_layer(4 * dim) def forward(self, x, H, W): """ Forward function. Args: x: Input feature, tensor size (B, H*W, C). H, W: Spatial resolution of the input feature. """ B, L, C = x.shape assert L == H * W, "input feature has wrong size" x = x.view(B, H, W, C) # padding pad_input = (H % 2 == 1) or (W % 2 == 1) if pad_input: x = F.pad(x, (0, 0, 0, W % 2, 0, H % 2)) x0 = x[:, 0::2, 0::2, :] # B H/2 W/2 C x1 = x[:, 1::2, 0::2, :] # B H/2 W/2 C x2 = x[:, 0::2, 1::2, :] # B H/2 W/2 C x3 = x[:, 1::2, 1::2, :] # B H/2 W/2 C x = torch.cat([x0, x1, x2, x3], -1) # B H/2 W/2 4*C x = x.view(B, -1, 4 * C) # B H/2*W/2 4*C x = self.norm(x) x = self.reduction(x) return x class BasicLayer(nn.Module): """ A basic Swin Transformer layer for one stage. Args: dim (int): Number of feature channels depth (int): Depths of this stage. num_heads (int): Number of attention head. window_size (int): Local window size. Default: 7. mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4. qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set. drop (float, optional): Dropout rate. Default: 0.0 attn_drop (float, optional): Attention dropout rate. Default: 0.0 drop_path (float | tuple[float], optional): Stochastic depth rate. Default: 0.0 norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm downsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default: None use_checkpoint (bool): Whether to use checkpointing to save memory. Default: False. """ def __init__(self, dim, depth, num_heads, window_size=7, mlp_ratio=4., qkv_bias=True, qk_scale=None, drop=0., attn_drop=0., drop_path=0., norm_layer=nn.LayerNorm, downsample=None, use_checkpoint=False): super().__init__() self.window_size = window_size self.shift_size = window_size // 2 self.depth = depth self.use_checkpoint = use_checkpoint # build blocks self.blocks = nn.ModuleList([ SwinTransformerBlock( dim=dim, num_heads=num_heads, window_size=window_size, shift_size=0 if (i % 2 == 0) else window_size // 2, mlp_ratio=mlp_ratio, qkv_bias=qkv_bias, qk_scale=qk_scale, drop=drop, attn_drop=attn_drop, drop_path=drop_path[i] if isinstance(drop_path, list) else drop_path, norm_layer=norm_layer) for i in range(depth)]) # patch merging layer if downsample is not None: self.downsample = downsample(dim=dim, norm_layer=norm_layer) else: self.downsample = None def forward(self, x, H, W): """ Forward function. Args: x: Input feature, tensor size (B, H*W, C). H, W: Spatial resolution of the input feature. """ # calculate attention mask for SW-MSA Hp = int(np.ceil(H / self.window_size)) * self.window_size Wp = int(np.ceil(W / self.window_size)) * self.window_size img_mask = torch.zeros((1, Hp, Wp, 1), device=x.device) # 1 Hp Wp 1 h_slices = (slice(0, -self.window_size), slice(-self.window_size, -self.shift_size), slice(-self.shift_size, None)) w_slices = (slice(0, -self.window_size), slice(-self.window_size, -self.shift_size), slice(-self.shift_size, None)) cnt = 0 for h in h_slices: for w in w_slices: img_mask[:, h, w, :] = cnt cnt += 1 mask_windows = window_partition(img_mask, self.window_size) # nW, window_size, window_size, 1 mask_windows = mask_windows.view(-1, self.window_size * self.window_size) attn_mask = mask_windows.unsqueeze(1) - mask_windows.unsqueeze(2) attn_mask = attn_mask.masked_fill(attn_mask != 0, float(-100.0)).masked_fill(attn_mask == 0, float(0.0)) for blk in self.blocks: blk.H, blk.W = H, W if self.use_checkpoint: x = checkpoint.checkpoint(blk, x, attn_mask) else: x = blk(x, attn_mask) if self.downsample is not None: x_down = self.downsample(x, H, W) Wh, Ww = (H + 1) // 2, (W + 1) // 2 return x, H, W, x_down, Wh, Ww else: return x, H, W, x, H, W class PatchEmbed(nn.Module): """ Image to Patch Embedding Args: patch_size (int): Patch token size. Default: 4. in_chans (int): Number of input image channels. Default: 3. embed_dim (int): Number of linear projection output channels. Default: 96. norm_layer (nn.Module, optional): Normalization layer. Default: None """ def __init__(self, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None): super().__init__() patch_size = to_2tuple(patch_size) self.patch_size = patch_size self.in_chans = in_chans self.embed_dim = embed_dim self.proj = nn.Conv2d(in_chans, embed_dim, kernel_size=patch_size, stride=patch_size) if norm_layer is not None: self.norm = norm_layer(embed_dim) else: self.norm = None def forward(self, x): """Forward function.""" # padding _, _, H, W = x.size() if W % self.patch_size[1] != 0: x = F.pad(x, (0, self.patch_size[1] - W % self.patch_size[1])) if H % self.patch_size[0] != 0: x = F.pad(x, (0, 0, 0, self.patch_size[0] - H % self.patch_size[0])) x = self.proj(x) # B C Wh Ww if self.norm is not None: Wh, Ww = x.size(2), x.size(3) x = x.flatten(2).transpose(1, 2) x = self.norm(x) x = x.transpose(1, 2).view(-1, self.embed_dim, Wh, Ww) return x #@BACKBONES.register_module() class swin(nn.Module): """ Swin Transformer backbone. A PyTorch impl of : `Swin Transformer: Hierarchical Vision Transformer using Shifted Windows` - https://arxiv.org/pdf/2103.14030 Args: pretrain_img_size (int): Input image size for training the pretrained model, used in absolute postion embedding. Default 224. patch_size (int | tuple(int)): Patch size. Default: 4. in_chans (int): Number of input image channels. Default: 3. embed_dim (int): Number of linear projection output channels. Default: 96. depths (tuple[int]): Depths of each Swin Transformer stage. num_heads (tuple[int]): Number of attention head of each stage. window_size (int): Window size. Default: 7. mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4. qkv_bias (bool): If True, add a learnable bias to query, key, value. Default: True qk_scale (float): Override default qk scale of head_dim ** -0.5 if set. drop_rate (float): Dropout rate. attn_drop_rate (float): Attention dropout rate. Default: 0. drop_path_rate (float): Stochastic depth rate. Default: 0.2. norm_layer (nn.Module): Normalization layer. Default: nn.LayerNorm. ape (bool): If True, add absolute position embedding to the patch embedding. Default: False. patch_norm (bool): If True, add normalization after patch embedding. Default: True. out_indices (Sequence[int]): Output from which stages. frozen_stages (int): Stages to be frozen (stop grad and set eval mode). -1 means not freezing any parameters. use_checkpoint (bool): Whether to use checkpointing to save memory. Default: False. """ def __init__(self, pretrain_img_size=224, patch_size=4, in_chans=3, embed_dim=96, depths=[2, 2, 6, 2], num_heads=[3, 6, 12, 24], window_size=7, mlp_ratio=4., qkv_bias=True, qk_scale=None, drop_rate=0., attn_drop_rate=0., drop_path_rate=0.2, norm_layer=nn.LayerNorm, ape=False, patch_norm=True, out_indices=(0, 1, 2, 3), frozen_stages=-1, use_checkpoint=False, pretrained=None, init_cfg=None): assert not (init_cfg and pretrained), \ 'init_cfg and pretrained cannot be specified at the same time' if isinstance(pretrained, str): warnings.warn('DeprecationWarning: pretrained is deprecated, ' 'please use "init_cfg" instead') init_cfg = dict(type='Pretrained', checkpoint=pretrained) elif pretrained is None: init_cfg = init_cfg else: raise TypeError('pretrained must be a str or None') super().__init__() self.pretrain_img_size = pretrain_img_size self.num_layers = len(depths) self.embed_dim = embed_dim self.ape = ape self.patch_norm = patch_norm self.out_indices = out_indices self.frozen_stages = frozen_stages self.init_cfg = init_cfg # split image into non-overlapping patches self.patch_embed = PatchEmbed( patch_size=patch_size, in_chans=in_chans, embed_dim=embed_dim, norm_layer=norm_layer if self.patch_norm else None) # absolute position embedding if self.ape: pretrain_img_size = to_2tuple(pretrain_img_size) patch_size = to_2tuple(patch_size) patches_resolution = [pretrain_img_size[0] // patch_size[0], pretrain_img_size[1] // patch_size[1]] self.absolute_pos_embed = nn.Parameter(torch.zeros(1, embed_dim, patches_resolution[0], patches_resolution[1])) trunc_normal_(self.absolute_pos_embed, std=.02) self.pos_drop = nn.Dropout(p=drop_rate) # stochastic depth dpr = [x.item() for x in torch.linspace(0, drop_path_rate, sum(depths))] # stochastic depth decay rule # build layers self.layers = nn.ModuleList() for i_layer in range(self.num_layers): layer = BasicLayer( dim=int(embed_dim * 2 ** i_layer), depth=depths[i_layer], num_heads=num_heads[i_layer], window_size=window_size, mlp_ratio=mlp_ratio, qkv_bias=qkv_bias, qk_scale=qk_scale, drop=drop_rate, attn_drop=attn_drop_rate, drop_path=dpr[sum(depths[:i_layer]):sum(depths[:i_layer + 1])], norm_layer=norm_layer, downsample=PatchMerging if (i_layer < self.num_layers - 1) else None, use_checkpoint=use_checkpoint) self.layers.append(layer) num_features = [int(embed_dim * 2 ** i) for i in range(self.num_layers)] self.num_features = num_features # add a norm layer for each output for i_layer in out_indices: layer = norm_layer(num_features[i_layer]) layer_name = f'norm{i_layer}' self.add_module(layer_name, layer) self._freeze_stages() self.out_channels = (3, 96, 192, 384, 768) def _freeze_stages(self): if self.frozen_stages >= 0: self.patch_embed.eval() for param in self.patch_embed.parameters(): param.requires_grad = False if self.frozen_stages >= 1 and self.ape: self.absolute_pos_embed.requires_grad = False if self.frozen_stages >= 2: self.pos_drop.eval() for i in range(0, self.frozen_stages - 1): m = self.layers[i] m.eval() for param in m.parameters(): param.requires_grad = False def init_weights(self, pretrained): ckpt = torch.load(pretrained, map_location='cpu') if 'state_dict' in ckpt: _state_dict = ckpt['state_dict'] elif 'model' in ckpt: _state_dict = ckpt['model'] else: _state_dict = ckpt state_dict = OrderedDict() for k, v in _state_dict.items(): if k.startswith('backbone.'): state_dict[k[9:]] = v else: state_dict[k] = v # strip prefix of state_dict if list(state_dict.keys())[0].startswith('module.'): state_dict = {k[7:]: v for k, v in state_dict.items()} # reshape absolute position embedding if state_dict.get('absolute_pos_embed') is not None: absolute_pos_embed = state_dict['absolute_pos_embed'] N1, L, C1 = absolute_pos_embed.size() N2, C2, H, W = self.absolute_pos_embed.size() if N1 != N2 or C1 != C2 or L != H * W: warnings.warn('Error in loading absolute_pos_embed, pass') else: state_dict['absolute_pos_embed'] = absolute_pos_embed.view( N2, H, W, C2).permute(0, 3, 1, 2).contiguous() # interpolate position bias table if needed relative_position_bias_table_keys = [ k for k in state_dict.keys() if 'relative_position_bias_table' in k ] for table_key in relative_position_bias_table_keys: table_pretrained = state_dict[table_key] table_current = self.state_dict()[table_key] L1, nH1 = table_pretrained.size() L2, nH2 = table_current.size() if nH1 != nH2: warnings.warn(f'Error in loading {table_key}, pass') elif L1 != L2: S1 = int(L1**0.5) S2 = int(L2**0.5) table_pretrained_resized = F.interpolate( table_pretrained.permute(1, 0).reshape(1, nH1, S1, S1), size=(S2, S2), mode='bicubic') state_dict[table_key] = table_pretrained_resized.view( nH2, L2).permute(1, 0).contiguous() # print('##############') # print(self.state_dict().keys()) # print('$$$$$$$$$$$$$$') # print(state_dict.keys()) # load state_dict msg = self.load_state_dict(state_dict, False) print(msg) def forward(self, x): """Forward function.""" img = [x] x = self.patch_embed(x) Wh, Ww = x.size(2), x.size(3) if self.ape: # interpolate the position embedding to the corresponding size absolute_pos_embed = F.interpolate(self.absolute_pos_embed, size=(Wh, Ww), mode='bicubic') x = (x + absolute_pos_embed).flatten(2).transpose(1, 2) # B Wh*Ww C else: x = x.flatten(2).transpose(1, 2) x = self.pos_drop(x) outs = [] for i in range(self.num_layers): layer = self.layers[i] x_out, H, W, x, Wh, Ww = layer(x, Wh, Ww) if i in self.out_indices: norm_layer = getattr(self, f'norm{i}') x_out = norm_layer(x_out) out = x_out.view(-1, H, W, self.num_features[i]).permute(0, 3, 1, 2).contiguous() outs.append(out) return img + outs def train(self, mode=True): """Convert the model into training mode while keep layers freezed.""" super(swin, self).train(mode) self._freeze_stages() ================================================ FILE: RStask/InstanceSegmentation/test.py ================================================ from RStask import InstanceFunction model=InstanceFunction('cuda:0') model.inference('/data/haonan.guo/LoveDA/Train/Urban/images_png/1367.png','bike','/data/haonan.guo/RSChatGPT/output.png') ================================================ FILE: RStask/InstanceSegmentation/uper.py ================================================ # Copyright (c) OpenMMLab. All rights reserved. import torch import torch.nn as nn import warnings from mmcv.cnn import ConvModule from abc import ABCMeta, abstractmethod from typing import List, Tuple import torch import torch.nn as nn from mmengine.model import BaseModule from torch import Tensor import torch.nn.functional as F def resize(input, size=None, scale_factor=None, mode='nearest', align_corners=None, warning=True): if warning: if size is not None and align_corners: input_h, input_w = tuple(int(x) for x in input.shape[2:]) output_h, output_w = tuple(int(x) for x in size) if output_h > input_h or output_w > output_h: if ((output_h > 1 and output_w > 1 and input_h > 1 and input_w > 1) and (output_h - 1) % (input_h - 1) and (output_w - 1) % (input_w - 1)): warnings.warn( f'When align_corners={align_corners}, ' 'the output would more aligned if ' f'input size {(input_h, input_w)} is `x+1` and ' f'out size {(output_h, output_w)} is `nx+1`') return F.interpolate(input, size, scale_factor, mode, align_corners) class PPM(nn.ModuleList): """Pooling Pyramid Module used in PSPNet. Args: pool_scales (tuple[int]): Pooling scales used in Pooling Pyramid Module. in_channels (int): Input channels. channels (int): Channels after modules, before conv_seg. conv_cfg (dict|None): Config of conv layers. norm_cfg (dict|None): Config of norm layers. act_cfg (dict): Config of activation layers. align_corners (bool): align_corners argument of F.interpolate. """ def __init__(self, pool_scales, in_channels, channels, conv_cfg, norm_cfg, act_cfg, align_corners, **kwargs): super().__init__() self.pool_scales = pool_scales self.align_corners = align_corners self.in_channels = in_channels self.channels = channels self.conv_cfg = conv_cfg self.norm_cfg = norm_cfg self.act_cfg = act_cfg for pool_scale in pool_scales: self.append( nn.Sequential( nn.AdaptiveAvgPool2d(pool_scale), ConvModule( self.in_channels, self.channels, 1, conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg, act_cfg=self.act_cfg, **kwargs))) def forward(self, x): """Forward function.""" ppm_outs = [] for ppm in self: ppm_out = ppm(x) upsampled_ppm_out = resize( ppm_out, size=x.size()[2:], mode='bilinear', align_corners=self.align_corners) ppm_outs.append(upsampled_ppm_out) return ppm_outs class BaseDecodeHead(BaseModule, metaclass=ABCMeta): """Base class for BaseDecodeHead. 1. The ``init_weights`` method is used to initialize decode_head's model parameters. After segmentor initialization, ``init_weights`` is triggered when ``segmentor.init_weights()`` is called externally. 2. The ``loss`` method is used to calculate the loss of decode_head, which includes two steps: (1) the decode_head model performs forward propagation to obtain the feature maps (2) The ``loss_by_feat`` method is called based on the feature maps to calculate the loss. .. code:: text loss(): forward() -> loss_by_feat() 3. The ``predict`` method is used to predict segmentation results, which includes two steps: (1) the decode_head model performs forward propagation to obtain the feature maps (2) The ``predict_by_feat`` method is called based on the feature maps to predict segmentation results including post-processing. .. code:: text predict(): forward() -> predict_by_feat() Args: in_channels (int|Sequence[int]): Input channels. channels (int): Channels after modules, before conv_seg. num_classes (int): Number of classes. out_channels (int): Output channels of conv_seg. threshold (float): Threshold for binary segmentation in the case of `num_classes==1`. Default: None. dropout_ratio (float): Ratio of dropout layer. Default: 0.1. conv_cfg (dict|None): Config of conv layers. Default: None. norm_cfg (dict|None): Config of norm layers. Default: None. act_cfg (dict): Config of activation layers. Default: dict(type='ReLU') in_index (int|Sequence[int]): Input feature index. Default: -1 input_transform (str|None): Transformation type of input features. Options: 'resize_concat', 'multiple_select', None. 'resize_concat': Multiple feature maps will be resize to the same size as first one and than concat together. Usually used in FCN head of HRNet. 'multiple_select': Multiple feature maps will be bundle into a list and passed into decode head. None: Only one select feature map is allowed. Default: None. loss_decode (dict | Sequence[dict]): Config of decode loss. The `loss_name` is property of corresponding loss function which could be shown in training log. If you want this loss item to be included into the backward graph, `loss_` must be the prefix of the name. Defaults to 'loss_ce'. e.g. dict(type='CrossEntropyLoss'), [dict(type='CrossEntropyLoss', loss_name='loss_ce'), dict(type='DiceLoss', loss_name='loss_dice')] Default: dict(type='CrossEntropyLoss'). ignore_index (int | None): The label index to be ignored. When using masked BCE loss, ignore_index should be set to None. Default: 255. sampler (dict|None): The config of segmentation map sampler. Default: None. align_corners (bool): align_corners argument of F.interpolate. Default: False. init_cfg (dict or list[dict], optional): Initialization config dict. """ def __init__(self, in_channels, channels, *, num_classes=None, out_channels=None, threshold=None, dropout_ratio=0.1, conv_cfg=None, norm_cfg=None, act_cfg=dict(type='ReLU'), in_index=-1, input_transform=None, loss_decode=dict( type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0), ignore_index=255, sampler=None, align_corners=False, init_cfg=dict( type='Normal', std=0.01, override=dict(name='conv_seg'))): super().__init__(init_cfg) self._init_inputs(in_channels, in_index, input_transform) self.channels = channels self.dropout_ratio = dropout_ratio self.conv_cfg = conv_cfg self.norm_cfg = norm_cfg self.act_cfg = act_cfg self.in_index = in_index self.ignore_index = ignore_index self.align_corners = align_corners if out_channels is None: if num_classes == 2: warnings.warn('For binary segmentation, we suggest using' '`out_channels = 1` to define the output' 'channels of segmentor, and use `threshold`' 'to convert `seg_logits` into a prediction' 'applying a threshold') out_channels = num_classes if out_channels != num_classes and out_channels != 1: raise ValueError( 'out_channels should be equal to num_classes,' 'except binary segmentation set out_channels == 1 and' f'num_classes == 2, but got out_channels={out_channels}' f'and num_classes={num_classes}') if out_channels == 1 and threshold is None: threshold = 0.3 warnings.warn('threshold is not defined for binary, and defaults' 'to 0.3') self.num_classes = num_classes self.out_channels = out_channels self.threshold = threshold # if isinstance(loss_decode, dict): # self.loss_decode = build_loss(loss_decode) # elif isinstance(loss_decode, (list, tuple)): # self.loss_decode = nn.ModuleList() # for loss in loss_decode: # self.loss_decode.append(build_loss(loss)) # else: # raise TypeError(f'loss_decode must be a dict or sequence of dict,\ # but got {type(loss_decode)}') if sampler is not None: self.sampler = build_pixel_sampler(sampler, context=self) else: self.sampler = None # self.conv_seg = nn.Conv2d(channels, self.out_channels, kernel_size=1) if dropout_ratio > 0: self.dropout = nn.Dropout2d(dropout_ratio) else: self.dropout = None def extra_repr(self): """Extra repr.""" s = f'input_transform={self.input_transform}, ' \ f'ignore_index={self.ignore_index}, ' \ f'align_corners={self.align_corners}' return s def _init_inputs(self, in_channels, in_index, input_transform): """Check and initialize input transforms. The in_channels, in_index and input_transform must match. Specifically, when input_transform is None, only single feature map will be selected. So in_channels and in_index must be of type int. When input_transform Args: in_channels (int|Sequence[int]): Input channels. in_index (int|Sequence[int]): Input feature index. input_transform (str|None): Transformation type of input features. Options: 'resize_concat', 'multiple_select', None. 'resize_concat': Multiple feature maps will be resize to the same size as first one and than concat together. Usually used in FCN head of HRNet. 'multiple_select': Multiple feature maps will be bundle into a list and passed into decode head. None: Only one select feature map is allowed. """ if input_transform is not None: assert input_transform in ['resize_concat', 'multiple_select'] self.input_transform = input_transform self.in_index = in_index if input_transform is not None: assert isinstance(in_channels, (list, tuple)) assert isinstance(in_index, (list, tuple)) assert len(in_channels) == len(in_index) if input_transform == 'resize_concat': self.in_channels = sum(in_channels) else: self.in_channels = in_channels else: assert isinstance(in_channels, int) assert isinstance(in_index, int) self.in_channels = in_channels def _transform_inputs(self, inputs): """Transform inputs for decoder. Args: inputs (list[Tensor]): List of multi-level img features. Returns: Tensor: The transformed inputs """ if self.input_transform == 'resize_concat': inputs = [inputs[i] for i in self.in_index] upsampled_inputs = [ resize( input=x, size=inputs[0].shape[2:], mode='bilinear', align_corners=self.align_corners) for x in inputs ] inputs = torch.cat(upsampled_inputs, dim=1) elif self.input_transform == 'multiple_select': inputs = [inputs[i] for i in self.in_index] else: inputs = inputs[self.in_index] return inputs @abstractmethod def forward(self, inputs): """Placeholder of forward function.""" pass def cls_seg(self, feat): """Classify each pixel.""" if self.dropout is not None: feat = self.dropout(feat) output = self.conv_seg(feat) return output def predict(self, inputs: Tuple[Tensor], batch_img_metas: List[dict], test_cfg) -> Tensor: """Forward function for prediction. Args: inputs (Tuple[Tensor]): List of multi-level img features. batch_img_metas (dict): List Image info where each dict may also contain: 'img_shape', 'scale_factor', 'flip', 'img_path', 'ori_shape', and 'pad_shape'. For details on the values of these keys see `mmseg/datasets/pipelines/formatting.py:PackSegInputs`. test_cfg (dict): The testing config. Returns: Tensor: Outputs segmentation logits map. """ seg_logits = self.forward(inputs) return self.predict_by_feat(seg_logits, batch_img_metas) def _stack_batch_gt(self, batch_data_samples) -> Tensor: gt_semantic_segs = [ data_sample.gt_sem_seg.data for data_sample in batch_data_samples ] return torch.stack(gt_semantic_segs, dim=0) def predict_by_feat(self, seg_logits: Tensor, batch_img_metas: List[dict]) -> Tensor: """Transform a batch of output seg_logits to the input shape. Args: seg_logits (Tensor): The output from decode head forward function. batch_img_metas (list[dict]): Meta information of each image, e.g., image size, scaling factor, etc. Returns: Tensor: Outputs segmentation logits map. """ seg_logits = resize( input=seg_logits, size=batch_img_metas[0]['img_shape'], mode='bilinear', align_corners=self.align_corners) return seg_logits #@MODELS.register_module() class UPerHead(BaseDecodeHead): """Unified Perceptual Parsing for Scene Understanding. This head is the implementation of `UPerNet `_. Args: pool_scales (tuple[int]): Pooling scales used in Pooling Pyramid Module applied on the last feature. Default: (1, 2, 3, 6). """ def __init__(self, pool_scales=(1, 2, 3, 6), **kwargs): super().__init__(input_transform='multiple_select', **kwargs) # PSP Module self.psp_modules = PPM( pool_scales, self.in_channels[-1], self.channels, conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg, act_cfg=self.act_cfg, align_corners=self.align_corners) self.bottleneck = ConvModule( self.in_channels[-1] + len(pool_scales) * self.channels, self.channels, 3, padding=1, conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg, act_cfg=self.act_cfg) # FPN Module self.lateral_convs = nn.ModuleList() self.fpn_convs = nn.ModuleList() for in_channels in self.in_channels[:-1]: # skip the top layer l_conv = ConvModule( in_channels, self.channels, 1, conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg, act_cfg=self.act_cfg, inplace=False) fpn_conv = ConvModule( self.channels, self.channels, 3, padding=1, conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg, act_cfg=self.act_cfg, inplace=False) self.lateral_convs.append(l_conv) self.fpn_convs.append(fpn_conv) self.fpn_bottleneck = ConvModule( len(self.in_channels) * self.channels, self.channels, 3, padding=1, conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg, act_cfg=self.act_cfg) def psp_forward(self, inputs): """Forward function of PSP module.""" x = inputs[-1] psp_outs = [x] psp_outs.extend(self.psp_modules(x)) psp_outs = torch.cat(psp_outs, dim=1) output = self.bottleneck(psp_outs) return output def _forward_feature(self, inputs): """Forward function for feature maps before classifying each pixel with ``self.cls_seg`` fc. Args: inputs (list[Tensor]): List of multi-level img features. Returns: feats (Tensor): A tensor of shape (batch_size, self.channels, H, W) which is feature map for last layer of decoder head. """ inputs = self._transform_inputs(inputs) # build laterals laterals = [ lateral_conv(inputs[i]) for i, lateral_conv in enumerate(self.lateral_convs) ] laterals.append(self.psp_forward(inputs)) # build top-down path used_backbone_levels = len(laterals) for i in range(used_backbone_levels - 1, 0, -1): prev_shape = laterals[i - 1].shape[2:] laterals[i - 1] = laterals[i - 1] + resize( laterals[i], size=prev_shape, mode='bilinear', align_corners=self.align_corners) # build outputs fpn_outs = [ self.fpn_convs[i](laterals[i]) for i in range(used_backbone_levels - 1) ] # append psp feature fpn_outs.append(laterals[-1]) for i in range(used_backbone_levels - 1, 0, -1): fpn_outs[i] = resize( fpn_outs[i], size=fpn_outs[0].shape[2:], mode='bilinear', align_corners=self.align_corners) fpn_outs = torch.cat(fpn_outs, dim=1) feats = self.fpn_bottleneck(fpn_outs) return feats def forward(self, *inputs): """Forward function.""" inputs = inputs[1:] output = self._forward_feature(inputs) #output = self.cls_seg(output) output = F.interpolate(output, scale_factor=4, mode='bilinear') return output ================================================ FILE: RStask/LanduseSegmentation/__init__.py ================================================ ================================================ FILE: RStask/LanduseSegmentation/seg_hrnet.py ================================================ import logging from skimage import io import torch import torch.nn as nn import torch._utils import torch.nn.functional as F from PIL import Image import numpy as np BatchNorm2d=nn.BatchNorm2d relu_inplace=True BN_MOMENTUM = 0.1 ALIGN_CORNERS = None def conv3x3(in_planes, out_planes, stride=1): """3x3 convolution with padding""" return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride, padding=1, bias=False) class BasicBlock(nn.Module): expansion = 1 def __init__(self, inplanes, planes, stride=1, downsample=None): super(BasicBlock, self).__init__() self.conv1 = conv3x3(inplanes, planes, stride) self.bn1 = BatchNorm2d(planes, momentum=BN_MOMENTUM) self.relu = nn.ReLU(inplace=relu_inplace) self.conv2 = conv3x3(planes, planes) self.bn2 = BatchNorm2d(planes, momentum=BN_MOMENTUM) self.downsample = downsample self.stride = stride def forward(self, x): residual = x out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) if self.downsample is not None: residual = self.downsample(x) out = out + residual out = self.relu(out) return out class Bottleneck(nn.Module): expansion = 4 def __init__(self, inplanes, planes, stride=1, downsample=None): super(Bottleneck, self).__init__() self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) self.bn1 = BatchNorm2d(planes, momentum=BN_MOMENTUM) self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride, padding=1, bias=False) self.bn2 = BatchNorm2d(planes, momentum=BN_MOMENTUM) self.conv3 = nn.Conv2d(planes, planes * self.expansion, kernel_size=1, bias=False) self.bn3 = BatchNorm2d(planes * self.expansion, momentum=BN_MOMENTUM) self.relu = nn.ReLU(inplace=relu_inplace) self.downsample = downsample self.stride = stride def forward(self, x): residual = x out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) out = self.relu(out) out = self.conv3(out) out = self.bn3(out) if self.downsample is not None: residual = self.downsample(x) out = out + residual out = self.relu(out) return out class HighResolutionModule(nn.Module): def __init__(self, num_branches, blocks, num_blocks, num_inchannels, num_channels, fuse_method, multi_scale_output=True): super(HighResolutionModule, self).__init__() self._check_branches( num_branches, blocks, num_blocks, num_inchannels, num_channels) self.num_inchannels = num_inchannels self.fuse_method = fuse_method self.num_branches = num_branches self.multi_scale_output = multi_scale_output self.branches = self._make_branches( num_branches, blocks, num_blocks, num_channels) self.fuse_layers = self._make_fuse_layers() self.relu = nn.ReLU(inplace=relu_inplace) def _check_branches(self, num_branches, blocks, num_blocks, num_inchannels, num_channels): if num_branches != len(num_blocks): error_msg = 'NUM_BRANCHES({}) <> NUM_BLOCKS({})'.format( num_branches, len(num_blocks)) raise ValueError(error_msg) if num_branches != len(num_channels): error_msg = 'NUM_BRANCHES({}) <> NUM_CHANNELS({})'.format( num_branches, len(num_channels)) raise ValueError(error_msg) if num_branches != len(num_inchannels): error_msg = 'NUM_BRANCHES({}) <> NUM_INCHANNELS({})'.format( num_branches, len(num_inchannels)) raise ValueError(error_msg) def _make_one_branch(self, branch_index, block, num_blocks, num_channels, stride=1): downsample = None if stride != 1 or \ self.num_inchannels[branch_index] != num_channels[branch_index] * block.expansion: downsample = nn.Sequential( nn.Conv2d(self.num_inchannels[branch_index], num_channels[branch_index] * block.expansion, kernel_size=1, stride=stride, bias=False), BatchNorm2d(num_channels[branch_index] * block.expansion, momentum=BN_MOMENTUM), ) layers = [] layers.append(block(self.num_inchannels[branch_index], num_channels[branch_index], stride, downsample)) self.num_inchannels[branch_index] = \ num_channels[branch_index] * block.expansion for i in range(1, num_blocks[branch_index]): layers.append(block(self.num_inchannels[branch_index], num_channels[branch_index])) return nn.Sequential(*layers) def _make_branches(self, num_branches, block, num_blocks, num_channels): branches = [] for i in range(num_branches): branches.append( self._make_one_branch(i, block, num_blocks, num_channels)) return nn.ModuleList(branches) def _make_fuse_layers(self): if self.num_branches == 1: return None num_branches = self.num_branches num_inchannels = self.num_inchannels fuse_layers = [] for i in range(num_branches if self.multi_scale_output else 1): fuse_layer = [] for j in range(num_branches): if j > i: fuse_layer.append(nn.Sequential( nn.Conv2d(num_inchannels[j], num_inchannels[i], 1, 1, 0, bias=False), BatchNorm2d(num_inchannels[i], momentum=BN_MOMENTUM))) elif j == i: fuse_layer.append(None) else: conv3x3s = [] for k in range(i-j): if k == i - j - 1: num_outchannels_conv3x3 = num_inchannels[i] conv3x3s.append(nn.Sequential( nn.Conv2d(num_inchannels[j], num_outchannels_conv3x3, 3, 2, 1, bias=False), BatchNorm2d(num_outchannels_conv3x3, momentum=BN_MOMENTUM))) else: num_outchannels_conv3x3 = num_inchannels[j] conv3x3s.append(nn.Sequential( nn.Conv2d(num_inchannels[j], num_outchannels_conv3x3, 3, 2, 1, bias=False), BatchNorm2d(num_outchannels_conv3x3, momentum=BN_MOMENTUM), nn.ReLU(inplace=relu_inplace))) fuse_layer.append(nn.Sequential(*conv3x3s)) fuse_layers.append(nn.ModuleList(fuse_layer)) return nn.ModuleList(fuse_layers) def get_num_inchannels(self): return self.num_inchannels def forward(self, x): if self.num_branches == 1: return [self.branches[0](x[0])] for i in range(self.num_branches): x[i] = self.branches[i](x[i]) x_fuse = [] for i in range(len(self.fuse_layers)): y = x[0] if i == 0 else self.fuse_layers[i][0](x[0]) for j in range(1, self.num_branches): if i == j: y = y + x[j] elif j > i: width_output = x[i].shape[-1] height_output = x[i].shape[-2] y = y + F.interpolate( self.fuse_layers[i][j](x[j]), size=[height_output, width_output], mode='bilinear', align_corners=ALIGN_CORNERS) else: y = y + self.fuse_layers[i][j](x[j]) x_fuse.append(self.relu(y)) return x_fuse blocks_dict = { 'BASIC': BasicBlock, 'BOTTLENECK': Bottleneck } class hrmodel(nn.Module): def __init__(self): super(hrmodel, self).__init__() BN_MOMENTUM,relu_inplace=0.1,True # stem net self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=2, padding=1, bias=False) self.bn1 = BatchNorm2d(64, momentum=BN_MOMENTUM) self.conv2 = nn.Conv2d(64, 64, kernel_size=3, stride=2, padding=1, bias=False) self.bn2 = BatchNorm2d(64, momentum=BN_MOMENTUM) self.relu = nn.ReLU(inplace=relu_inplace) self.stage1_cfg = {'NUM_MODULES': 1, 'NUM_RANCHES': 1, 'BLOCK': 'BOTTLENECK', 'NUM_BLOCKS': [4], 'NUM_CHANNELS': [64], 'FUSE_METHOD': 'SUM'} num_channels = self.stage1_cfg['NUM_CHANNELS'][0] block = blocks_dict[self.stage1_cfg['BLOCK']] num_blocks = self.stage1_cfg['NUM_BLOCKS'][0] self.layer1 = self._make_layer(block, 64, 64, 4) stage1_out_channel = block.expansion*num_channels self.stage2_cfg = {'NUM_MODULES': 1, 'NUM_BRANCHES': 2, 'BLOCK': 'BASIC', 'NUM_BLOCKS': [4, 4], 'NUM_CHANNELS': [48, 96], 'FUSE_METHOD': 'SUM'} num_channels = self.stage2_cfg['NUM_CHANNELS'] block = blocks_dict[self.stage2_cfg['BLOCK']] num_channels = [ num_channels[i] * block.expansion for i in range(len(num_channels))] self.transition1 = self._make_transition_layer( [stage1_out_channel], num_channels) self.stage2, pre_stage_channels = self._make_stage( self.stage2_cfg, num_channels) self.stage3_cfg = {'NUM_MODULES': 4, 'NUM_BRANCHES': 3, 'BLOCK': 'BASIC', 'NUM_BLOCKS': [4, 4, 4], 'NUM_CHANNELS': [48, 96, 192], 'FUSE_METHOD': 'SUM'} num_channels = self.stage3_cfg['NUM_CHANNELS'] block = blocks_dict[self.stage3_cfg['BLOCK']] num_channels = [ num_channels[i] * block.expansion for i in range(len(num_channels))] self.transition2 = self._make_transition_layer( pre_stage_channels, num_channels) self.stage3, pre_stage_channels = self._make_stage( self.stage3_cfg, num_channels) self.stage4_cfg = {'NUM_MODULES': 3, 'NUM_BRANCHES': 4, 'BLOCK': 'BASIC', 'NUM_BLOCKS': [4, 4, 4, 4], 'NUM_CHANNELS': [48, 96, 192, 384], 'FUSE_METHOD': 'SUM'} num_channels = self.stage4_cfg['NUM_CHANNELS'] block = blocks_dict[self.stage4_cfg['BLOCK']] num_channels = [ num_channels[i] * block.expansion for i in range(len(num_channels))] self.transition3 = self._make_transition_layer( pre_stage_channels, num_channels) self.stage4, pre_stage_channels = self._make_stage( self.stage4_cfg, num_channels, multi_scale_output=True) self.convs=nn.Sequential(ConvModule()) self.conv_seg=nn.Conv2d(720, 7, kernel_size=(1, 1), stride=(1, 1)) def _make_transition_layer( self, num_channels_pre_layer, num_channels_cur_layer): num_branches_cur = len(num_channels_cur_layer) num_branches_pre = len(num_channels_pre_layer) transition_layers = [] for i in range(num_branches_cur): if i < num_branches_pre: if num_channels_cur_layer[i] != num_channels_pre_layer[i]: transition_layers.append(nn.Sequential( nn.Conv2d(num_channels_pre_layer[i], num_channels_cur_layer[i], 3, 1, 1, bias=False), BatchNorm2d( num_channels_cur_layer[i], momentum=BN_MOMENTUM), nn.ReLU(inplace=relu_inplace))) else: transition_layers.append(None) else: conv3x3s = [] for j in range(i+1-num_branches_pre): inchannels = num_channels_pre_layer[-1] outchannels = num_channels_cur_layer[i] \ if j == i-num_branches_pre else inchannels conv3x3s.append(nn.Sequential( nn.Conv2d( inchannels, outchannels, 3, 2, 1, bias=False), BatchNorm2d(outchannels, momentum=BN_MOMENTUM), nn.ReLU(inplace=relu_inplace))) transition_layers.append(nn.Sequential(*conv3x3s)) return nn.ModuleList(transition_layers) def _make_layer(self, block, inplanes, planes, blocks, stride=1): downsample = None if stride != 1 or inplanes != planes * block.expansion: downsample = nn.Sequential( nn.Conv2d(inplanes, planes * block.expansion, kernel_size=1, stride=stride, bias=False), BatchNorm2d(planes * block.expansion, momentum=BN_MOMENTUM), ) layers = [] layers.append(block(inplanes, planes, stride, downsample)) inplanes = planes * block.expansion for i in range(1, blocks): layers.append(block(inplanes, planes)) return nn.Sequential(*layers) def _make_stage(self, layer_config, num_inchannels, multi_scale_output=True): num_modules = layer_config['NUM_MODULES'] num_branches = layer_config['NUM_BRANCHES'] num_blocks = layer_config['NUM_BLOCKS'] num_channels = layer_config['NUM_CHANNELS'] block = blocks_dict[layer_config['BLOCK']] fuse_method = layer_config['FUSE_METHOD'] modules = [] for i in range(num_modules): # multi_scale_output is only used last module if not multi_scale_output and i == num_modules - 1: reset_multi_scale_output = False else: reset_multi_scale_output = True modules.append( HighResolutionModule(num_branches, block, num_blocks, num_inchannels, num_channels, fuse_method, reset_multi_scale_output) ) num_inchannels = modules[-1].get_num_inchannels() return nn.Sequential(*modules), num_inchannels def forward(self, x,gts=None): x = self.conv1(x) x = self.bn1(x) x = self.relu(x) x = self.conv2(x) x = self.bn2(x) x = self.relu(x) x = self.layer1(x) x_list = [] for i in range(self.stage2_cfg['NUM_BRANCHES']): if self.transition1[i] is not None: x_list.append(self.transition1[i](x)) else: x_list.append(x) y_list = self.stage2(x_list) x_list = [] for i in range(self.stage3_cfg['NUM_BRANCHES']): if self.transition2[i] is not None: if i < self.stage2_cfg['NUM_BRANCHES']: x_list.append(self.transition2[i](y_list[i])) else: x_list.append(self.transition2[i](y_list[-1])) else: x_list.append(y_list[i]) y_list = self.stage3(x_list) x_list = [] for i in range(self.stage4_cfg['NUM_BRANCHES']): if self.transition3[i] is not None: if i < self.stage3_cfg['NUM_BRANCHES']: x_list.append(self.transition3[i](y_list[i])) else: x_list.append(self.transition3[i](y_list[-1])) else: x_list.append(y_list[i]) x = self.stage4(x_list) # return x # Upsampling x0_h, x0_w = x[0].size(2), x[0].size(3) x1 = F.interpolate(x[1], size=(x0_h, x0_w), mode='bilinear', align_corners=ALIGN_CORNERS) x2 = F.interpolate(x[2], size=(x0_h, x0_w), mode='bilinear', align_corners=ALIGN_CORNERS) x3 = F.interpolate(x[3], size=(x0_h, x0_w), mode='bilinear', align_corners=ALIGN_CORNERS) x = torch.cat([x[0], x1, x2, x3], 1) mid=x # x = self.last_layer(x) x=self.convs(x) x=self.conv_seg(x) return x class ConvModule(nn.Module): def __init__(self): super(ConvModule, self).__init__() self.conv=nn.Conv2d(720, 720, kernel_size=(1, 1), stride=(1, 1), bias=False) self.bn=nn.BatchNorm2d(720, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) self.relu=nn.ReLU(inplace=True) def forward(self,x): return self.relu(self.bn(self.conv(x))) class HRNet48(nn.Module): def __init__(self,device): super(HRNet48, self).__init__() self.model=hrmodel() self.device = device try: trained = torch.load('./checkpoints/HRNET_LoveDA_best.pth') except: trained = torch.load('../../checkpoints/HRNET_LoveDA_best.pth') self.load_state_dict(trained) self.model = self.model.to(device) self.model.eval() self.category = ['Background','Building', 'Road', 'Water', 'Barren', 'Forest', 'Farmland'] self.color_bar=[[0,0,0],[255,0,0],[255,255,0],[0,0,255],[128,0,128],[0,255,0],[255,128,0]] self.mean, self.std = torch.tensor([123.675, 116.28, 103.53]).reshape((1, 3, 1, 1)), torch.tensor( [58.395, 57.12, 57.375]).reshape((1, 3, 1, 1)) def visualize(self,pred,cls): vis=np.zeros([pred.shape[0],pred.shape[1],3]).astype(np.uint8) if len(cls)>1: for i in range(len(self.category)): vis[:,:,0][pred==i]=self.color_bar[i][0] vis[:,:,1][pred == i] = self.color_bar[i][1] vis[:,:,2][pred == i] = self.color_bar[i][2] else: idx=cls[0] vis[:, :, 0][pred == idx] = self.color_bar[idx][0] vis[:, :, 1][pred == idx] = self.color_bar[idx][1] vis[:, :, 2][pred == idx] = self.color_bar[idx][2] return vis def inference(self,image_path, det_prompt,updated_image_path): det_prompt=det_prompt.strip() image = torch.from_numpy(io.imread(image_path)) image = (image.permute(2, 0, 1).unsqueeze(0) - self.mean) / self.std with torch.no_grad(): b, c, h, w = image.shape pred = self.model(image.to(self.device)) pred = F.interpolate(pred, (h, w), mode='bilinear') pred = pred.argmax(1).cpu().squeeze().int().numpy() if det_prompt.lower() == 'landuse': pred_vis = self.visualize(pred, self.category) elif det_prompt.lower() in [i.lower() for i in self.category]: idx=[i.lower() for i in self.category].index(det_prompt.strip().lower()) pred_vis = self.visualize(pred, [idx]) else: print('Category ',det_prompt,' do not suuport!') return ('Category ',det_prompt,' do not suuport!','The expected input category include Building, Road, Water, Barren, Forest, Farmland, Landuse.') pred = Image.fromarray(pred_vis.astype(np.uint8)) pred.save(updated_image_path) print(f"\nProcessed Landuse Segmentation, Input Image: {image_path+','+det_prompt}, Output: {updated_image_path}") return det_prompt+' segmentation result in '+updated_image_path if __name__=='__main__': net=HRNet48() print(sum(p.numel() for p in net.parameters())) x=torch.ones((2,3,512,512)) output=net(x) print(output.shape) ================================================ FILE: RStask/LanduseSegmentation/test.py ================================================ from RStask import LanduseFunction model=LanduseFunction('cuda:0') model.inference('/data/haonan.guo/LoveDA/Train/Urban/images_png/1367.png','road','/data/haonan.guo/RSChatGPT/output.png') ================================================ FILE: RStask/ObjectCounting/Yolocounting.py ================================================ from RStask.ObjectDetection.models.common import DetectMultiBackend import torch from skimage import io import numpy as np import torchvision import torch.nn.functional as F class YoloCounting: def __init__(self, device): from RStask.ObjectDetection.models.common import DetectMultiBackend self.device = device try: self.model = DetectMultiBackend('./checkpoints/yolov5_best.pt', device=torch.device(device), dnn=False, fp16=False) except: self.model = DetectMultiBackend('../../checkpoints/yolov5_best.pt', device=torch.device(device), dnn=False,fp16=False) self.category = ['small vehicle', 'large vehicle', 'plane', 'storage tank', 'ship', 'harbor', 'ground track field', 'soccer ball field', 'tennis court', 'swimming pool', 'baseball diamond', 'roundabout', 'basketball court', 'bridge', 'helicopter'] def inference(self, image_path, det_prompt): supported_class=False for i in range(len(self.category)): if self.category[i] == det_prompt or self.category[i] == det_prompt[:-1] or self.category[i] == det_prompt[:-3]: supported_class=True if supported_class is False: log_text=det_prompt+' is not a supported category for the model.' print(f"\nProcessed Object Counting, Input Image: {image_path}, Output text: {log_text}") return log_text image = torch.from_numpy(io.imread(image_path)) image = image.permute(2, 0, 1).unsqueeze(0) / 255.0 _, _, h, w = image.shape with torch.no_grad(): out, _ = self.model(image.to(self.device), augment=False,val=True) predn = self.non_max_suppression(out, conf_thres=0.001, iou_thres=0.75, labels=[], multi_label=True, agnostic=False)[0] detections = predn.clone() detections = detections[predn[:, 4] > 0.75] detections_box = (detections[:, :4] / (640 / h)).int().cpu().numpy() detection_classes = detections[:, 5].int().cpu().numpy() log_text = '' for i in range(len(self.category)): if (detection_classes == i).sum() > 0 and ( self.category[i] == det_prompt or self.category[i] == det_prompt[:-1] or self.category[ i] == det_prompt[:-3]): log_text += str((detection_classes == i).sum()) + ' ' + self.category[i] + ',' if log_text != '': log_text = log_text[:-1] + ' detected.' else: log_text = 'No ' + self.category[i] + ' detected.' print(f"\nProcessed Object Counting, Input Image: {image_path}, Output text: {log_text}") return log_text def non_max_suppression(self, prediction, conf_thres=0.25, iou_thres=0.45, classes=None, agnostic=False, multi_label=False, labels=(), max_det=300): """Non-Maximum Suppression (NMS) on inference results to reject overlapping bounding boxes Returns: list of detections, on (n,6) tensor per image [xyxy, conf, cls] """ def box_iou(box1, box2): def box_area(box): # box = xyxy(4,n) return (box[2] - box[0]) * (box[3] - box[1]) # https://github.com/pytorch/vision/blob/master/torchvision/ops/boxes.py """ Return intersection-over-union (Jaccard index) of boxes. Both sets of boxes are expected to be in (x1, y1, x2, y2) format. Arguments: box1 (Tensor[N, 4]) box2 (Tensor[M, 4]) Returns: iou (Tensor[N, M]): the NxM matrix containing the pairwise IoU values for every element in boxes1 and boxes2 """ # inter(N,M) = (rb(N,M,2) - lt(N,M,2)).clamp(0).prod(2) (a1, a2), (b1, b2) = box1[:, None].chunk(2, 2), box2.chunk(2, 1) inter = (torch.min(a2, b2) - torch.max(a1, b1)).clamp(0).prod(2) # IoU = inter / (area1 + area2 - inter) return inter / (box_area(box1.T)[:, None] + box_area(box2.T) - inter) def xywh2xyxy(x): # Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right y = x.clone() y[:, 0] = x[:, 0] - x[:, 2] / 2 # top left x y[:, 1] = x[:, 1] - x[:, 3] / 2 # top left y y[:, 2] = x[:, 0] + x[:, 2] / 2 # bottom right x y[:, 3] = x[:, 1] + x[:, 3] / 2 # bottom right y return y bs = prediction.shape[0] # batch size nc = prediction.shape[2] - 5 # number of classes xc = prediction[..., 4] > conf_thres # candidates # Checks assert 0 <= conf_thres <= 1, f'Invalid Confidence threshold {conf_thres}, valid values are between 0.0 and 1.0' assert 0 <= iou_thres <= 1, f'Invalid IoU {iou_thres}, valid values are between 0.0 and 1.0' # Settings # min_wh = 2 # (pixels) minimum box width and height max_wh = 7680 # (pixels) maximum box width and height max_nms = 30000 # maximum number of boxes into torchvision.ops.nms() time_limit = 0.1 + 0.03 * bs # seconds to quit after redundant = True # require redundant detections multi_label &= nc > 1 # multiple labels per box (adds 0.5ms/img) merge = False # use merge-NMS output = [torch.zeros((0, 6), device=prediction.device)] * bs for xi, x in enumerate(prediction): # image index, image inference # Apply constraints # x[((x[..., 2:4] < min_wh) | (x[..., 2:4] > max_wh)).any(1), 4] = 0 # width-height x = x[xc[xi]] # confidence # Cat apriori labels if autolabelling if labels and len(labels[xi]): lb = labels[xi] v = torch.zeros((len(lb), nc + 5), device=x.device) v[:, :4] = lb[:, 1:5] # box v[:, 4] = 1.0 # conf v[range(len(lb)), lb[:, 0].long() + 5] = 1.0 # cls x = torch.cat((x, v), 0) # If none remain process next image if not x.shape[0]: continue # Compute conf x[:, 5:] *= x[:, 4:5] # conf = obj_conf * cls_conf # Box (center x, center y, width, height) to (x1, y1, x2, y2) box = xywh2xyxy(x[:, :4]) # Detections matrix nx6 (xyxy, conf, cls) if multi_label: i, j = (x[:, 5:] > conf_thres).nonzero(as_tuple=False).T x = torch.cat((box[i], x[i, j + 5, None], j[:, None].float()), 1) else: # best class only conf, j = x[:, 5:].max(1, keepdim=True) x = torch.cat((box, conf, j.float()), 1)[conf.view(-1) > conf_thres] # Filter by class if classes is not None: x = x[(x[:, 5:6] == torch.tensor(classes, device=x.device)).any(1)] # Apply finite constraint # if not torch.isfinite(x).all(): # x = x[torch.isfinite(x).all(1)] # Check shape n = x.shape[0] # number of boxes if not n: # no boxes continue elif n > max_nms: # excess boxes x = x[x[:, 4].argsort(descending=True)[:max_nms]] # sort by confidence # Batched NMS c = x[:, 5:6] * (0 if agnostic else max_wh) # classes boxes, scores = x[:, :4] + c, x[:, 4] # boxes (offset by class), scores i = torchvision.ops.nms(boxes, scores, iou_thres) # NMS if i.shape[0] > max_det: # limit detections i = i[:max_det] if merge and (1 < n < 3E3): # Merge NMS (boxes merged using weighted mean) # update boxes as boxes(i,4) = weights(i,n) * boxes(n,4) iou = box_iou(boxes[i], boxes) > iou_thres # iou matrix weights = iou * scores[None] # box weights x[i, :4] = torch.mm(weights, x[:, :4]).float() / weights.sum(1, keepdim=True) # merged boxes if redundant: i = i[iou.sum(1) > 1] # require redundancy output[xi] = x[i] return output ================================================ FILE: RStask/ObjectCounting/__init__.py ================================================ ================================================ FILE: RStask/ObjectCounting/test.py ================================================ from RStask import CountingFuncnction model=CountingFuncnction('cuda:0') txt='/data/haonan.guo/RSChatGPT/test.tif,small vehicles' p,t=txt.split(",") model.inference(p,t) ================================================ FILE: RStask/ObjectDetection/YOLOv5.py ================================================ from RStask.ObjectDetection.models.common import DetectMultiBackend import torch from skimage import io import numpy as np import torchvision import cv2 from PIL import Image class YoloDetection: def __init__(self, device): self.device = device try: self.model = DetectMultiBackend('./checkpoints/yolov5_best.pt', device=torch.device(device), dnn=False, fp16=False) except: self.model = DetectMultiBackend('/data/haonan.guo/RSChatGPT/checkpoints/yolov5_best.pt', device=torch.device(device), dnn=False,fp16=False) self.category = ['small vehicle', 'large vehicle', 'plane', 'storage tank', 'ship', 'harbor', 'ground track field', 'soccer ball field', 'tennis court', 'swimming pool', 'baseball diamond', 'roundabout', 'basketball court', 'bridge', 'helicopter'] def inference(self, image_path, det_prompt,updated_image_path): image = torch.from_numpy(io.imread(image_path)) image = image.permute(2, 0, 1).unsqueeze(0) / 255.0 _, _, h, w = image.shape with torch.no_grad(): out, _ = self.model(image.to(self.device), augment=False,val=True) predn = self.non_max_suppression(out, conf_thres=0.001, iou_thres=0.75, labels=[], multi_label=True, agnostic=False)[0] detections = predn.clone() detections = detections[predn[:, 4] > 0.75] detections_box = (detections[:, :4] / (640 / h)).int().cpu().numpy() detection_classes = detections[:, 5].int().cpu().numpy() if len(detection_classes) > 0: det = np.zeros((h, w, 3)) for i in range(len(detections_box)): x1, y1, x2, y2 = detections_box[i] det[y1:y2, x1:x2] = detection_classes[i] + 1 self.visualize(image_path,updated_image_path,detections) print( f"\nProcessed Object Detection, Input Image: {image_path}, Output Bounding box: {updated_image_path},Output text: {'Object Detection Done'}") return det_prompt+' object detection result in '+updated_image_path def visualize(self,image_path, newpic_path,detections): font = cv2.FONT_HERSHEY_SIMPLEX im = io.imread(image_path) boxes = detections.int().cpu().numpy() for i in range(len(boxes)): cv2.rectangle(im, (boxes[i][0], boxes[i][1]), (boxes[i][2], boxes[i][3]), (0, 255, 255), 2) cv2.rectangle(im, (boxes[i][0], boxes[i][1] - 15), (boxes[i][0] + 45, boxes[i][1] - 2), (0, 0, 255),thickness=-1) cv2.putText(im, self.category[boxes[i][-1]], (boxes[i][0], boxes[i][1] - 2), font, 0.5, (255, 255, 255),1) Image.fromarray(im.astype(np.uint8)).save(newpic_path) with open(newpic_path[:-4]+'.txt','w') as f: for i in range(len(boxes)): f.write(str(list(boxes[i,:4]))[1:-1]+', '+self.category[boxes[i][-1]]+'\n') def non_max_suppression(self, prediction, conf_thres=0.25, iou_thres=0.45, classes=None, agnostic=False, multi_label=False, labels=(), max_det=300): """Non-Maximum Suppression (NMS) on inference results to reject overlapping bounding boxes Returns: list of detections, on (n,6) tensor per image [xyxy, conf, cls] """ def box_iou(box1, box2): def box_area(box): # box = xyxy(4,n) return (box[2] - box[0]) * (box[3] - box[1]) # https://github.com/pytorch/vision/blob/master/torchvision/ops/boxes.py """ Return intersection-over-union (Jaccard index) of boxes. Both sets of boxes are expected to be in (x1, y1, x2, y2) format. Arguments: box1 (Tensor[N, 4]) box2 (Tensor[M, 4]) Returns: iou (Tensor[N, M]): the NxM matrix containing the pairwise IoU values for every element in boxes1 and boxes2 """ # inter(N,M) = (rb(N,M,2) - lt(N,M,2)).clamp(0).prod(2) (a1, a2), (b1, b2) = box1[:, None].chunk(2, 2), box2.chunk(2, 1) inter = (torch.min(a2, b2) - torch.max(a1, b1)).clamp(0).prod(2) # IoU = inter / (area1 + area2 - inter) return inter / (box_area(box1.T)[:, None] + box_area(box2.T) - inter) def xywh2xyxy(x): # Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right y = x.clone() y[:, 0] = x[:, 0] - x[:, 2] / 2 # top left x y[:, 1] = x[:, 1] - x[:, 3] / 2 # top left y y[:, 2] = x[:, 0] + x[:, 2] / 2 # bottom right x y[:, 3] = x[:, 1] + x[:, 3] / 2 # bottom right y return y bs = prediction.shape[0] # batch size nc = prediction.shape[2] - 5 # number of classes xc = prediction[..., 4] > conf_thres # candidates # Checks assert 0 <= conf_thres <= 1, f'Invalid Confidence threshold {conf_thres}, valid values are between 0.0 and 1.0' assert 0 <= iou_thres <= 1, f'Invalid IoU {iou_thres}, valid values are between 0.0 and 1.0' # Settings # min_wh = 2 # (pixels) minimum box width and height max_wh = 7680 # (pixels) maximum box width and height max_nms = 30000 # maximum number of boxes into torchvision.ops.nms() time_limit = 0.1 + 0.03 * bs # seconds to quit after redundant = True # require redundant detections multi_label &= nc > 1 # multiple labels per box (adds 0.5ms/img) merge = False # use merge-NMS output = [torch.zeros((0, 6), device=prediction.device)] * bs for xi, x in enumerate(prediction): # image index, image inference # Apply constraints # x[((x[..., 2:4] < min_wh) | (x[..., 2:4] > max_wh)).any(1), 4] = 0 # width-height x = x[xc[xi]] # confidence # Cat apriori labels if autolabelling if labels and len(labels[xi]): lb = labels[xi] v = torch.zeros((len(lb), nc + 5), device=x.device) v[:, :4] = lb[:, 1:5] # box v[:, 4] = 1.0 # conf v[range(len(lb)), lb[:, 0].long() + 5] = 1.0 # cls x = torch.cat((x, v), 0) # If none remain process next image if not x.shape[0]: continue # Compute conf x[:, 5:] *= x[:, 4:5] # conf = obj_conf * cls_conf # Box (center x, center y, width, height) to (x1, y1, x2, y2) box = xywh2xyxy(x[:, :4]) # Detections matrix nx6 (xyxy, conf, cls) if multi_label: i, j = (x[:, 5:] > conf_thres).nonzero(as_tuple=False).T x = torch.cat((box[i], x[i, j + 5, None], j[:, None].float()), 1) else: # best class only conf, j = x[:, 5:].max(1, keepdim=True) x = torch.cat((box, conf, j.float()), 1)[conf.view(-1) > conf_thres] # Filter by class if classes is not None: x = x[(x[:, 5:6] == torch.tensor(classes, device=x.device)).any(1)] # Apply finite constraint # if not torch.isfinite(x).all(): # x = x[torch.isfinite(x).all(1)] # Check shape n = x.shape[0] # number of boxes if not n: # no boxes continue elif n > max_nms: # excess boxes x = x[x[:, 4].argsort(descending=True)[:max_nms]] # sort by confidence # Batched NMS c = x[:, 5:6] * (0 if agnostic else max_wh) # classes boxes, scores = x[:, :4] + c, x[:, 4] # boxes (offset by class), scores i = torchvision.ops.nms(boxes, scores, iou_thres) # NMS if i.shape[0] > max_det: # limit detections i = i[:max_det] if merge and (1 < n < 3E3): # Merge NMS (boxes merged using weighted mean) # update boxes as boxes(i,4) = weights(i,n) * boxes(n,4) iou = box_iou(boxes[i], boxes) > iou_thres # iou matrix weights = iou * scores[None] # box weights x[i, :4] = torch.mm(weights, x[:, :4]).float() / weights.sum(1, keepdim=True) # merged boxes if redundant: i = i[iou.sum(1) > 1] # require redundancy output[xi] = x[i] return output ================================================ FILE: RStask/ObjectDetection/__init__.py ================================================ ================================================ FILE: RStask/ObjectDetection/models/__init__.py ================================================ ================================================ FILE: RStask/ObjectDetection/models/common.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Common modules """ import json import math import platform import warnings from collections import OrderedDict, namedtuple from copy import copy from pathlib import Path import cv2 import numpy as np import pandas as pd import requests import torch import torch.nn as nn import yaml from PIL import Image from torch.cuda import amp from RStask.ObjectDetection.utils.datasets import exif_transpose, letterbox from RStask.ObjectDetection.utils.general import (LOGGER, check_requirements, check_suffix, check_version, colorstr, increment_path, make_divisible, non_max_suppression, scale_coords, xywh2xyxy, xyxy2xywh) from RStask.ObjectDetection.utils.plots import Annotator, colors, save_one_box from RStask.ObjectDetection.utils.torch_utils import copy_attr, time_sync def autopad(k, p=None): # kernel, padding # Pad to 'same' if p is None: p = k // 2 if isinstance(k, int) else (x // 2 for x in k) # auto-pad return p class Conv(nn.Module): # Standard convolution def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups super().__init__() self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False) self.bn = nn.BatchNorm2d(c2) self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity()) def forward(self, x): return self.act(self.bn(self.conv(x))) def forward_fuse(self, x): return self.act(self.conv(x)) class DWConv(Conv): # Depth-wise convolution class def __init__(self, c1, c2, k=1, s=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups super().__init__(c1, c2, k, s, g=math.gcd(c1, c2), act=act) class TransformerLayer(nn.Module): # Transformer layer https://arxiv.org/abs/2010.11929 (LayerNorm layers removed for better performance) def __init__(self, c, num_heads): super().__init__() self.q = nn.Linear(c, c, bias=False) self.k = nn.Linear(c, c, bias=False) self.v = nn.Linear(c, c, bias=False) self.ma = nn.MultiheadAttention(embed_dim=c, num_heads=num_heads) self.fc1 = nn.Linear(c, c, bias=False) self.fc2 = nn.Linear(c, c, bias=False) def forward(self, x): x = self.ma(self.q(x), self.k(x), self.v(x))[0] + x x = self.fc2(self.fc1(x)) + x return x class TransformerBlock(nn.Module): # Vision Transformer https://arxiv.org/abs/2010.11929 def __init__(self, c1, c2, num_heads, num_layers): super().__init__() self.conv = None if c1 != c2: self.conv = Conv(c1, c2) self.linear = nn.Linear(c2, c2) # learnable position embedding self.tr = nn.Sequential(*(TransformerLayer(c2, num_heads) for _ in range(num_layers))) self.c2 = c2 def forward(self, x): if self.conv is not None: x = self.conv(x) b, _, w, h = x.shape p = x.flatten(2).permute(2, 0, 1) return self.tr(p + self.linear(p)).permute(1, 2, 0).reshape(b, self.c2, w, h) class Bottleneck(nn.Module): # Standard bottleneck def __init__(self, c1, c2, shortcut=True, g=1, e=0.5): # ch_in, ch_out, shortcut, groups, expansion super().__init__() c_ = int(c2 * e) # hidden channels self.cv1 = Conv(c1, c_, 1, 1) self.cv2 = Conv(c_, c2, 3, 1, g=g) self.add = shortcut and c1 == c2 def forward(self, x): return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x)) class BottleneckCSP(nn.Module): # CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion super().__init__() c_ = int(c2 * e) # hidden channels self.cv1 = Conv(c1, c_, 1, 1) self.cv2 = nn.Conv2d(c1, c_, 1, 1, bias=False) self.cv3 = nn.Conv2d(c_, c_, 1, 1, bias=False) self.cv4 = Conv(2 * c_, c2, 1, 1) self.bn = nn.BatchNorm2d(2 * c_) # applied to cat(cv2, cv3) self.act = nn.SiLU() self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n))) def forward(self, x): y1 = self.cv3(self.m(self.cv1(x))) y2 = self.cv2(x) return self.cv4(self.act(self.bn(torch.cat((y1, y2), 1)))) class C3(nn.Module): # CSP Bottleneck with 3 convolutions def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion super().__init__() c_ = int(c2 * e) # hidden channels self.cv1 = Conv(c1, c_, 1, 1) self.cv2 = Conv(c1, c_, 1, 1) self.cv3 = Conv(2 * c_, c2, 1) # optional act=FReLU(c2) self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n))) # self.m = nn.Sequential(*(CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n))) def forward(self, x): return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), 1)) class C3TR(C3): # C3 module with TransformerBlock() def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): super().__init__(c1, c2, n, shortcut, g, e) c_ = int(c2 * e) self.m = TransformerBlock(c_, c_, 4, n) class C3SPP(C3): # C3 module with SPP() def __init__(self, c1, c2, k=(5, 9, 13), n=1, shortcut=True, g=1, e=0.5): super().__init__(c1, c2, n, shortcut, g, e) c_ = int(c2 * e) self.m = SPP(c_, c_, k) class C3Ghost(C3): # C3 module with GhostBottleneck() def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): super().__init__(c1, c2, n, shortcut, g, e) c_ = int(c2 * e) # hidden channels self.m = nn.Sequential(*(GhostBottleneck(c_, c_) for _ in range(n))) class SPP(nn.Module): # Spatial Pyramid Pooling (SPP) layer https://arxiv.org/abs/1406.4729 def __init__(self, c1, c2, k=(5, 9, 13)): super().__init__() c_ = c1 // 2 # hidden channels self.cv1 = Conv(c1, c_, 1, 1) self.cv2 = Conv(c_ * (len(k) + 1), c2, 1, 1) self.m = nn.ModuleList([nn.MaxPool2d(kernel_size=x, stride=1, padding=x // 2) for x in k]) def forward(self, x): x = self.cv1(x) with warnings.catch_warnings(): warnings.simplefilter('ignore') # suppress torch 1.9.0 max_pool2d() warning return self.cv2(torch.cat([x] + [m(x) for m in self.m], 1)) class SPPF(nn.Module): # Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv5 by Glenn Jocher def __init__(self, c1, c2, k=5): # equivalent to SPP(k=(5, 9, 13)) super().__init__() c_ = c1 // 2 # hidden channels self.cv1 = Conv(c1, c_, 1, 1) self.cv2 = Conv(c_ * 4, c2, 1, 1) self.m = nn.MaxPool2d(kernel_size=k, stride=1, padding=k // 2) def forward(self, x): x = self.cv1(x) with warnings.catch_warnings(): warnings.simplefilter('ignore') # suppress torch 1.9.0 max_pool2d() warning y1 = self.m(x) y2 = self.m(y1) return self.cv2(torch.cat((x, y1, y2, self.m(y2)), 1)) class Focus(nn.Module): # Focus wh information into c-space def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups super().__init__() self.conv = Conv(c1 * 4, c2, k, s, p, g, act) # self.contract = Contract(gain=2) def forward(self, x): # x(b,c,w,h) -> y(b,4c,w/2,h/2) return self.conv(torch.cat((x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]), 1)) # return self.conv(self.contract(x)) class GhostConv(nn.Module): # Ghost Convolution https://github.com/huawei-noah/ghostnet def __init__(self, c1, c2, k=1, s=1, g=1, act=True): # ch_in, ch_out, kernel, stride, groups super().__init__() c_ = c2 // 2 # hidden channels self.cv1 = Conv(c1, c_, k, s, None, g, act) self.cv2 = Conv(c_, c_, 5, 1, None, c_, act) def forward(self, x): y = self.cv1(x) return torch.cat((y, self.cv2(y)), 1) class GhostBottleneck(nn.Module): # Ghost Bottleneck https://github.com/huawei-noah/ghostnet def __init__(self, c1, c2, k=3, s=1): # ch_in, ch_out, kernel, stride super().__init__() c_ = c2 // 2 self.conv = nn.Sequential( GhostConv(c1, c_, 1, 1), # pw DWConv(c_, c_, k, s, act=False) if s == 2 else nn.Identity(), # dw GhostConv(c_, c2, 1, 1, act=False)) # pw-linear self.shortcut = nn.Sequential(DWConv(c1, c1, k, s, act=False), Conv(c1, c2, 1, 1, act=False)) if s == 2 else nn.Identity() def forward(self, x): return self.conv(x) + self.shortcut(x) class Contract(nn.Module): # Contract width-height into channels, i.e. x(1,64,80,80) to x(1,256,40,40) def __init__(self, gain=2): super().__init__() self.gain = gain def forward(self, x): b, c, h, w = x.size() # assert (h / s == 0) and (W / s == 0), 'Indivisible gain' s = self.gain x = x.view(b, c, h // s, s, w // s, s) # x(1,64,40,2,40,2) x = x.permute(0, 3, 5, 1, 2, 4).contiguous() # x(1,2,2,64,40,40) return x.view(b, c * s * s, h // s, w // s) # x(1,256,40,40) class Expand(nn.Module): # Expand channels into width-height, i.e. x(1,64,80,80) to x(1,16,160,160) def __init__(self, gain=2): super().__init__() self.gain = gain def forward(self, x): b, c, h, w = x.size() # assert C / s ** 2 == 0, 'Indivisible gain' s = self.gain x = x.view(b, s, s, c // s ** 2, h, w) # x(1,2,2,16,80,80) x = x.permute(0, 3, 4, 1, 5, 2).contiguous() # x(1,16,80,2,80,2) return x.view(b, c // s ** 2, h * s, w * s) # x(1,16,160,160) class Concat(nn.Module): # Concatenate a list of tensors along dimension def __init__(self, dimension=1): super().__init__() self.d = dimension def forward(self, x): return torch.cat(x, self.d) class DetectMultiBackend(nn.Module): # YOLOv5 MultiBackend class for python inference on various backends def __init__(self, weights='yolov5s.pt', device=torch.device('cpu'), dnn=False,fp16=False): from RStask.ObjectDetection.models.experimental import attempt_download, attempt_load # scoped to avoid circular import super().__init__() w = str(weights[0] if isinstance(weights, list) else weights) stride, names = 32, [f'class{i}' for i in range(1000)] # assign defaults w = attempt_download(w) # download if not local fp16 = False names = ['small-vehicle', 'large-vehicle','plane','storage-tank', 'ship','harbor','ground-track-field','soccer-ball-field', 'tennis-court','swimming-pool','baseball-diamond', 'roundabout','basketball-court','bridge','helicopter'] model = attempt_load(weights if isinstance(weights, list) else w, map_location=device) stride = max(int(model.stride.max()), 32) # model stride names = model.module.names if hasattr(model, 'module') else model.names # get class names model.half() if fp16 else model.float() self.model = model # explicitly assign for to(), cpu(), cuda(), half() self.__dict__.update(locals()) # assign all variables to self def forward(self, im, augment=False, visualize=False, val=False): # YOLOv5 MultiBackend inference b, ch, h, w = im.shape # batch, channel, height, width y = self.model(im, augment=augment, visualize=visualize)[0] if isinstance(y, np.ndarray): y = torch.tensor(y, device=self.device) return (y, []) if val else y class AutoShape(nn.Module): # YOLOv5 input-robust model wrapper for passing cv2/np/PIL/torch inputs. Includes preprocessing, inference and NMS conf = 0.25 # NMS confidence threshold iou = 0.45 # NMS IoU threshold agnostic = False # NMS class-agnostic multi_label = False # NMS multiple labels per box classes = None # (optional list) filter by class, i.e. = [0, 15, 16] for COCO persons, cats and dogs max_det = 1000 # maximum number of detections per image amp = False # Automatic Mixed Precision (AMP) inference def __init__(self, model): super().__init__() LOGGER.info('Adding AutoShape... ') copy_attr(self, model, include=('yaml', 'nc', 'hyp', 'names', 'stride', 'abc'), exclude=()) # copy attributes self.dmb = isinstance(model, DetectMultiBackend) # DetectMultiBackend() instance self.pt = not self.dmb or model.pt # PyTorch model self.model = model.eval() def _apply(self, fn): # Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers self = super()._apply(fn) if self.pt: m = self.model.model.model[-1] if self.dmb else self.model.model[-1] # Detect() m.stride = fn(m.stride) m.grid = list(map(fn, m.grid)) if isinstance(m.anchor_grid, list): m.anchor_grid = list(map(fn, m.anchor_grid)) return self @torch.no_grad() def forward(self, imgs, size=640, augment=False, profile=False): # Inference from various sources. For height=640, width=1280, RGB images example inputs are: # file: imgs = 'data/images/zidane.jpg' # str or PosixPath # URI: = 'https://ultralytics.com/images/zidane.jpg' # OpenCV: = cv2.imread('image.jpg')[:,:,::-1] # HWC BGR to RGB x(640,1280,3) # PIL: = Image.open('image.jpg') or ImageGrab.grab() # HWC x(640,1280,3) # numpy: = np.zeros((640,1280,3)) # HWC # torch: = torch.zeros(16,3,320,640) # BCHW (scaled to size=640, 0-1 values) # multiple: = [Image.open('image1.jpg'), Image.open('image2.jpg'), ...] # list of images t = [time_sync()] p = next(self.model.parameters()) if self.pt else torch.zeros(1, device=self.model.device) # for device, type autocast = self.amp and (p.device.type != 'cpu') # Automatic Mixed Precision (AMP) inference if isinstance(imgs, torch.Tensor): # torch with amp.autocast(autocast): return self.model(imgs.to(p.device).type_as(p), augment, profile) # inference # Pre-process n, imgs = (len(imgs), list(imgs)) if isinstance(imgs, (list, tuple)) else (1, [imgs]) # number, list of images shape0, shape1, files = [], [], [] # image and inference shapes, filenames for i, im in enumerate(imgs): f = f'image{i}' # filename if isinstance(im, (str, Path)): # filename or uri im, f = Image.open(requests.get(im, stream=True).raw if str(im).startswith('http') else im), im im = np.asarray(exif_transpose(im)) elif isinstance(im, Image.Image): # PIL Image im, f = np.asarray(exif_transpose(im)), getattr(im, 'filename', f) or f files.append(Path(f).with_suffix('.jpg').name) if im.shape[0] < 5: # image in CHW im = im.transpose((1, 2, 0)) # reverse dataloader .transpose(2, 0, 1) im = im[..., :3] if im.ndim == 3 else np.tile(im[..., None], 3) # enforce 3ch input s = im.shape[:2] # HWC shape0.append(s) # image shape g = (size / max(s)) # gain shape1.append([y * g for y in s]) imgs[i] = im if im.data.contiguous else np.ascontiguousarray(im) # update shape1 = [make_divisible(x, self.stride) if self.pt else size for x in np.array(shape1).max(0)] # inf shape x = [letterbox(im, shape1, auto=False)[0] for im in imgs] # pad x = np.ascontiguousarray(np.array(x).transpose((0, 3, 1, 2))) # stack and BHWC to BCHW x = torch.from_numpy(x).to(p.device).type_as(p) / 255 # uint8 to fp16/32 t.append(time_sync()) with amp.autocast(autocast): # Inference y = self.model(x, augment, profile) # forward t.append(time_sync()) # Post-process y = non_max_suppression(y if self.dmb else y[0], self.conf, self.iou, self.classes, self.agnostic, self.multi_label, max_det=self.max_det) # NMS for i in range(n): scale_coords(shape1, y[i][:, :4], shape0[i]) t.append(time_sync()) return Detections(imgs, y, files, t, self.names, x.shape) class Detections: # YOLOv5 detections class for inference results def __init__(self, imgs, pred, files, times=(0, 0, 0, 0), names=None, shape=None): super().__init__() d = pred[0].device # device gn = [torch.tensor([*(im.shape[i] for i in [1, 0, 1, 0]), 1, 1], device=d) for im in imgs] # normalizations self.imgs = imgs # list of images as numpy arrays self.pred = pred # list of tensors pred[0] = (xyxy, conf, cls) self.names = names # class names self.files = files # image filenames self.times = times # profiling times self.xyxy = pred # xyxy pixels self.xywh = [xyxy2xywh(x) for x in pred] # xywh pixels self.xyxyn = [x / g for x, g in zip(self.xyxy, gn)] # xyxy normalized self.xywhn = [x / g for x, g in zip(self.xywh, gn)] # xywh normalized self.n = len(self.pred) # number of images (batch size) self.t = tuple((times[i + 1] - times[i]) * 1000 / self.n for i in range(3)) # timestamps (ms) self.s = shape # inference BCHW shape def display(self, pprint=False, show=False, save=False, crop=False, render=False, labels=True, save_dir=Path('')): crops = [] for i, (im, pred) in enumerate(zip(self.imgs, self.pred)): s = f'image {i + 1}/{len(self.pred)}: {im.shape[0]}x{im.shape[1]} ' # string if pred.shape[0]: for c in pred[:, -1].unique(): n = (pred[:, -1] == c).sum() # detections per class s += f"{n} {self.names[int(c)]}{'s' * (n > 1)}, " # add to string if show or save or render or crop: annotator = Annotator(im, example=str(self.names)) for *box, conf, cls in reversed(pred): # xyxy, confidence, class label = f'{self.names[int(cls)]} {conf:.2f}' if crop: file = save_dir / 'crops' / self.names[int(cls)] / self.files[i] if save else None crops.append({ 'box': box, 'conf': conf, 'cls': cls, 'label': label, 'im': save_one_box(box, im, file=file, save=save)}) else: # all others annotator.box_label(box, label if labels else '', color=colors(cls)) im = annotator.im else: s += '(no detections)' im = Image.fromarray(im.astype(np.uint8)) if isinstance(im, np.ndarray) else im # from np if pprint: print(s.rstrip(', ')) if show: im.show(self.files[i]) # show if save: f = self.files[i] im.save(save_dir / f) # save if i == self.n - 1: LOGGER.info(f"Saved {self.n} image{'s' * (self.n > 1)} to {colorstr('bold', save_dir)}") if render: self.imgs[i] = np.asarray(im) if crop: if save: LOGGER.info(f'Saved results to {save_dir}\n') return crops def print(self): self.display(pprint=True) # print results print(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {tuple(self.s)}' % self.t) def show(self, labels=True): self.display(show=True, labels=labels) # show results def save(self, labels=True, save_dir='runs/detect/exp'): save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/detect/exp', mkdir=True) # increment save_dir self.display(save=True, labels=labels, save_dir=save_dir) # save results def crop(self, save=True, save_dir='runs/detect/exp'): save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/detect/exp', mkdir=True) if save else None return self.display(crop=True, save=save, save_dir=save_dir) # crop results def render(self, labels=True): self.display(render=True, labels=labels) # render results return self.imgs def pandas(self): # return detections as pandas DataFrames, i.e. print(results.pandas().xyxy[0]) new = copy(self) # return copy ca = 'xmin', 'ymin', 'xmax', 'ymax', 'confidence', 'class', 'name' # xyxy columns cb = 'xcenter', 'ycenter', 'width', 'height', 'confidence', 'class', 'name' # xywh columns for k, c in zip(['xyxy', 'xyxyn', 'xywh', 'xywhn'], [ca, ca, cb, cb]): a = [[x[:5] + [int(x[5]), self.names[int(x[5])]] for x in x.tolist()] for x in getattr(self, k)] # update setattr(new, k, [pd.DataFrame(x, columns=c) for x in a]) return new def tolist(self): # return a list of Detections objects, i.e. 'for result in results.tolist():' r = range(self.n) # iterable x = [Detections([self.imgs[i]], [self.pred[i]], [self.files[i]], self.times, self.names, self.s) for i in r] # for d in x: # for k in ['imgs', 'pred', 'xyxy', 'xyxyn', 'xywh', 'xywhn']: # setattr(d, k, getattr(d, k)[0]) # pop out of list return x def __len__(self): return self.n # override len(results) def __str__(self): self.print() # override print(results) return '' class Classify(nn.Module): # Classification head, i.e. x(b,c1,20,20) to x(b,c2) def __init__(self, c1, c2, k=1, s=1, p=None, g=1): # ch_in, ch_out, kernel, stride, padding, groups super().__init__() self.aap = nn.AdaptiveAvgPool2d(1) # to x(b,c1,1,1) self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g) # to x(b,c2,1,1) self.flat = nn.Flatten() def forward(self, x): z = torch.cat([self.aap(y) for y in (x if isinstance(x, list) else [x])], 1) # cat if list return self.flat(self.conv(z)) # flatten to x(b,c2) ================================================ FILE: RStask/ObjectDetection/models/experimental.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Experimental modules """ import math import numpy as np import torch import torch.nn as nn from RStask.ObjectDetection.models.common import Conv from RStask.ObjectDetection.utils.downloads import attempt_download class CrossConv(nn.Module): # Cross Convolution Downsample def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False): # ch_in, ch_out, kernel, stride, groups, expansion, shortcut super().__init__() c_ = int(c2 * e) # hidden channels self.cv1 = Conv(c1, c_, (1, k), (1, s)) self.cv2 = Conv(c_, c2, (k, 1), (s, 1), g=g) self.add = shortcut and c1 == c2 def forward(self, x): return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x)) class Sum(nn.Module): # Weighted sum of 2 or more layers https://arxiv.org/abs/1911.09070 def __init__(self, n, weight=False): # n: number of inputs super().__init__() self.weight = weight # apply weights boolean self.iter = range(n - 1) # iter object if weight: self.w = nn.Parameter(-torch.arange(1.0, n) / 2, requires_grad=True) # layer weights def forward(self, x): y = x[0] # no weight if self.weight: w = torch.sigmoid(self.w) * 2 for i in self.iter: y = y + x[i + 1] * w[i] else: for i in self.iter: y = y + x[i + 1] return y class MixConv2d(nn.Module): # Mixed Depth-wise Conv https://arxiv.org/abs/1907.09595 def __init__(self, c1, c2, k=(1, 3), s=1, equal_ch=True): # ch_in, ch_out, kernel, stride, ch_strategy super().__init__() n = len(k) # number of convolutions if equal_ch: # equal c_ per group i = torch.linspace(0, n - 1E-6, c2).floor() # c2 indices c_ = [(i == g).sum() for g in range(n)] # intermediate channels else: # equal weight.numel() per group b = [c2] + [0] * n a = np.eye(n + 1, n, k=-1) a -= np.roll(a, 1, axis=1) a *= np.array(k) ** 2 a[0] = 1 c_ = np.linalg.lstsq(a, b, rcond=None)[0].round() # solve for equal weight indices, ax = b self.m = nn.ModuleList([ nn.Conv2d(c1, int(c_), k, s, k // 2, groups=math.gcd(c1, int(c_)), bias=False) for k, c_ in zip(k, c_)]) self.bn = nn.BatchNorm2d(c2) self.act = nn.SiLU() def forward(self, x): return self.act(self.bn(torch.cat([m(x) for m in self.m], 1))) class Ensemble(nn.ModuleList): # Ensemble of models def __init__(self): super().__init__() def forward(self, x, augment=False, profile=False, visualize=False): y = [] for module in self: y.append(module(x, augment, profile, visualize)[0]) # y = torch.stack(y).max(0)[0] # max ensemble # y = torch.stack(y).mean(0) # mean ensemble y = torch.cat(y, 1) # nms ensemble return y, None # inference, train output def attempt_load(weights, map_location=None, inplace=True, fuse=True): from RStask.ObjectDetection.models.yolo import Detect, Model # Loads an ensemble of models weights=[a,b,c] or a single model weights=[a] or weights=a model = Ensemble() for w in weights if isinstance(weights, list) else [weights]: ckpt = torch.load(attempt_download(w), map_location=map_location) # load ckpt = (ckpt.get('ema') or ckpt['model']).float() # FP32 model model.append(ckpt.fuse().eval() if fuse else ckpt.eval()) # fused or un-fused model in eval mode # Compatibility updates for m in model.modules(): t = type(m) if t in (nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU, Detect, Model): m.inplace = inplace # torch 1.7.0 compatibility if t is Detect: if not isinstance(m.anchor_grid, list): # new Detect Layer compatibility delattr(m, 'anchor_grid') setattr(m, 'anchor_grid', [torch.zeros(1)] * m.nl) elif t is Conv: m._non_persistent_buffers_set = set() # torch 1.6.0 compatibility elif t is nn.Upsample and not hasattr(m, 'recompute_scale_factor'): m.recompute_scale_factor = None # torch 1.11.0 compatibility if len(model) == 1: return model[-1] # return model else: print(f'Ensemble created with {weights}\n') for k in 'names', 'nc', 'yaml': setattr(model, k, getattr(model[0], k)) model.stride = model[torch.argmax(torch.tensor([m.stride.max() for m in model])).int()].stride # max stride assert all(model[0].nc == m.nc for m in model), f'Models have different class counts: {[m.nc for m in model]}' return model # return ensemble ================================================ FILE: RStask/ObjectDetection/models/hub/anchors.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Default anchors for COCO data # P5 ------------------------------------------------------------------------------------------------------------------- # P5-640: anchors_p5_640: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # P6 ------------------------------------------------------------------------------------------------------------------- # P6-640: thr=0.25: 0.9964 BPR, 5.54 anchors past thr, n=12, img_size=640, metric_all=0.281/0.716-mean/best, past_thr=0.469-mean: 9,11, 21,19, 17,41, 43,32, 39,70, 86,64, 65,131, 134,130, 120,265, 282,180, 247,354, 512,387 anchors_p6_640: - [9,11, 21,19, 17,41] # P3/8 - [43,32, 39,70, 86,64] # P4/16 - [65,131, 134,130, 120,265] # P5/32 - [282,180, 247,354, 512,387] # P6/64 # P6-1280: thr=0.25: 0.9950 BPR, 5.55 anchors past thr, n=12, img_size=1280, metric_all=0.281/0.714-mean/best, past_thr=0.468-mean: 19,27, 44,40, 38,94, 96,68, 86,152, 180,137, 140,301, 303,264, 238,542, 436,615, 739,380, 925,792 anchors_p6_1280: - [19,27, 44,40, 38,94] # P3/8 - [96,68, 86,152, 180,137] # P4/16 - [140,301, 303,264, 238,542] # P5/32 - [436,615, 739,380, 925,792] # P6/64 # P6-1920: thr=0.25: 0.9950 BPR, 5.55 anchors past thr, n=12, img_size=1920, metric_all=0.281/0.714-mean/best, past_thr=0.468-mean: 28,41, 67,59, 57,141, 144,103, 129,227, 270,205, 209,452, 455,396, 358,812, 653,922, 1109,570, 1387,1187 anchors_p6_1920: - [28,41, 67,59, 57,141] # P3/8 - [144,103, 129,227, 270,205] # P4/16 - [209,452, 455,396, 358,812] # P5/32 - [653,922, 1109,570, 1387,1187] # P6/64 # P7 ------------------------------------------------------------------------------------------------------------------- # P7-640: thr=0.25: 0.9962 BPR, 6.76 anchors past thr, n=15, img_size=640, metric_all=0.275/0.733-mean/best, past_thr=0.466-mean: 11,11, 13,30, 29,20, 30,46, 61,38, 39,92, 78,80, 146,66, 79,163, 149,150, 321,143, 157,303, 257,402, 359,290, 524,372 anchors_p7_640: - [11,11, 13,30, 29,20] # P3/8 - [30,46, 61,38, 39,92] # P4/16 - [78,80, 146,66, 79,163] # P5/32 - [149,150, 321,143, 157,303] # P6/64 - [257,402, 359,290, 524,372] # P7/128 # P7-1280: thr=0.25: 0.9968 BPR, 6.71 anchors past thr, n=15, img_size=1280, metric_all=0.273/0.732-mean/best, past_thr=0.463-mean: 19,22, 54,36, 32,77, 70,83, 138,71, 75,173, 165,159, 148,334, 375,151, 334,317, 251,626, 499,474, 750,326, 534,814, 1079,818 anchors_p7_1280: - [19,22, 54,36, 32,77] # P3/8 - [70,83, 138,71, 75,173] # P4/16 - [165,159, 148,334, 375,151] # P5/32 - [334,317, 251,626, 499,474] # P6/64 - [750,326, 534,814, 1079,818] # P7/128 # P7-1920: thr=0.25: 0.9968 BPR, 6.71 anchors past thr, n=15, img_size=1920, metric_all=0.273/0.732-mean/best, past_thr=0.463-mean: 29,34, 81,55, 47,115, 105,124, 207,107, 113,259, 247,238, 222,500, 563,227, 501,476, 376,939, 749,711, 1126,489, 801,1222, 1618,1227 anchors_p7_1920: - [29,34, 81,55, 47,115] # P3/8 - [105,124, 207,107, 113,259] # P4/16 - [247,238, 222,500, 563,227] # P5/32 - [501,476, 376,939, 749,711] # P6/64 - [1126,489, 801,1222, 1618,1227] # P7/128 ================================================ FILE: RStask/ObjectDetection/models/hub/yolov3-spp.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # darknet53 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [32, 3, 1]], # 0 [-1, 1, Conv, [64, 3, 2]], # 1-P1/2 [-1, 1, Bottleneck, [64]], [-1, 1, Conv, [128, 3, 2]], # 3-P2/4 [-1, 2, Bottleneck, [128]], [-1, 1, Conv, [256, 3, 2]], # 5-P3/8 [-1, 8, Bottleneck, [256]], [-1, 1, Conv, [512, 3, 2]], # 7-P4/16 [-1, 8, Bottleneck, [512]], [-1, 1, Conv, [1024, 3, 2]], # 9-P5/32 [-1, 4, Bottleneck, [1024]], # 10 ] # YOLOv3-SPP head head: [[-1, 1, Bottleneck, [1024, False]], [-1, 1, SPP, [512, [5, 9, 13]]], [-1, 1, Conv, [1024, 3, 1]], [-1, 1, Conv, [512, 1, 1]], [-1, 1, Conv, [1024, 3, 1]], # 15 (P5/32-large) [-2, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P4 [-1, 1, Bottleneck, [512, False]], [-1, 1, Bottleneck, [512, False]], [-1, 1, Conv, [256, 1, 1]], [-1, 1, Conv, [512, 3, 1]], # 22 (P4/16-medium) [-2, 1, Conv, [128, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P3 [-1, 1, Bottleneck, [256, False]], [-1, 2, Bottleneck, [256, False]], # 27 (P3/8-small) [[27, 22, 15], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov3-tiny.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: - [10,14, 23,27, 37,58] # P4/16 - [81,82, 135,169, 344,319] # P5/32 # YOLOv3-tiny backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [16, 3, 1]], # 0 [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 1-P1/2 [-1, 1, Conv, [32, 3, 1]], [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 3-P2/4 [-1, 1, Conv, [64, 3, 1]], [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 5-P3/8 [-1, 1, Conv, [128, 3, 1]], [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 7-P4/16 [-1, 1, Conv, [256, 3, 1]], [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 9-P5/32 [-1, 1, Conv, [512, 3, 1]], [-1, 1, nn.ZeroPad2d, [[0, 1, 0, 1]]], # 11 [-1, 1, nn.MaxPool2d, [2, 1, 0]], # 12 ] # YOLOv3-tiny head head: [[-1, 1, Conv, [1024, 3, 1]], [-1, 1, Conv, [256, 1, 1]], [-1, 1, Conv, [512, 3, 1]], # 15 (P5/32-large) [-2, 1, Conv, [128, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P4 [-1, 1, Conv, [256, 3, 1]], # 19 (P4/16-medium) [[19, 15], 1, Detect, [nc, anchors]], # Detect(P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov3.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # darknet53 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [32, 3, 1]], # 0 [-1, 1, Conv, [64, 3, 2]], # 1-P1/2 [-1, 1, Bottleneck, [64]], [-1, 1, Conv, [128, 3, 2]], # 3-P2/4 [-1, 2, Bottleneck, [128]], [-1, 1, Conv, [256, 3, 2]], # 5-P3/8 [-1, 8, Bottleneck, [256]], [-1, 1, Conv, [512, 3, 2]], # 7-P4/16 [-1, 8, Bottleneck, [512]], [-1, 1, Conv, [1024, 3, 2]], # 9-P5/32 [-1, 4, Bottleneck, [1024]], # 10 ] # YOLOv3 head head: [[-1, 1, Bottleneck, [1024, False]], [-1, 1, Conv, [512, 1, 1]], [-1, 1, Conv, [1024, 3, 1]], [-1, 1, Conv, [512, 1, 1]], [-1, 1, Conv, [1024, 3, 1]], # 15 (P5/32-large) [-2, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P4 [-1, 1, Bottleneck, [512, False]], [-1, 1, Bottleneck, [512, False]], [-1, 1, Conv, [256, 1, 1]], [-1, 1, Conv, [512, 3, 1]], # 22 (P4/16-medium) [-2, 1, Conv, [128, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P3 [-1, 1, Bottleneck, [256, False]], [-1, 2, Bottleneck, [256, False]], # 27 (P3/8-small) [[27, 22, 15], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5-bifpn.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 BiFPN head head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 14, 6], 1, Concat, [1]], # cat P4 <--- BiFPN change [-1, 3, C3, [512, False]], # 20 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 23 (P5/32-large) [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5-fpn.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 FPN head head: [[-1, 3, C3, [1024, False]], # 10 (P5/32-large) [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 1, Conv, [512, 1, 1]], [-1, 3, C3, [512, False]], # 14 (P4/16-medium) [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 1, Conv, [256, 1, 1]], [-1, 3, C3, [256, False]], # 18 (P3/8-small) [[18, 14, 10], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5-p2.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: 3 # AutoAnchor evolves 3 anchors per P output layer # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 head with (P2, P3, P4, P5) outputs head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small) [-1, 1, Conv, [128, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 2], 1, Concat, [1]], # cat backbone P2 [-1, 1, C3, [128, False]], # 21 (P2/4-xsmall) [-1, 1, Conv, [128, 3, 2]], [[-1, 18], 1, Concat, [1]], # cat head P3 [-1, 3, C3, [256, False]], # 24 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 27 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 30 (P5/32-large) [[21, 24, 27, 30], 1, Detect, [nc, anchors]], # Detect(P2, P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5-p34.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 0.33 # model depth multiple width_multiple: 0.50 # layer channel multiple anchors: 3 # AutoAnchor evolves 3 anchors per P output layer # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [ [ -1, 1, Conv, [ 64, 6, 2, 2 ] ], # 0-P1/2 [ -1, 1, Conv, [ 128, 3, 2 ] ], # 1-P2/4 [ -1, 3, C3, [ 128 ] ], [ -1, 1, Conv, [ 256, 3, 2 ] ], # 3-P3/8 [ -1, 6, C3, [ 256 ] ], [ -1, 1, Conv, [ 512, 3, 2 ] ], # 5-P4/16 [ -1, 9, C3, [ 512 ] ], [ -1, 1, Conv, [ 1024, 3, 2 ] ], # 7-P5/32 [ -1, 3, C3, [ 1024 ] ], [ -1, 1, SPPF, [ 1024, 5 ] ], # 9 ] # YOLOv5 v6.0 head with (P3, P4) outputs head: [ [ -1, 1, Conv, [ 512, 1, 1 ] ], [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], [ [ -1, 6 ], 1, Concat, [ 1 ] ], # cat backbone P4 [ -1, 3, C3, [ 512, False ] ], # 13 [ -1, 1, Conv, [ 256, 1, 1 ] ], [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], [ [ -1, 4 ], 1, Concat, [ 1 ] ], # cat backbone P3 [ -1, 3, C3, [ 256, False ] ], # 17 (P3/8-small) [ -1, 1, Conv, [ 256, 3, 2 ] ], [ [ -1, 14 ], 1, Concat, [ 1 ] ], # cat head P4 [ -1, 3, C3, [ 512, False ] ], # 20 (P4/16-medium) [ [ 17, 20 ], 1, Detect, [ nc, anchors ] ], # Detect(P3, P4) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5-p6.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: 3 # AutoAnchor evolves 3 anchors per P output layer # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 [-1, 3, C3, [768]], [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 11 ] # YOLOv5 v6.0 head with (P3, P4, P5, P6) outputs head: [[-1, 1, Conv, [768, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P5 [-1, 3, C3, [768, False]], # 15 [-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 19 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 23 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 20], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 26 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 16], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [768, False]], # 29 (P5/32-large) [-1, 1, Conv, [768, 3, 2]], [[-1, 12], 1, Concat, [1]], # cat head P6 [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5-p7.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: 3 # AutoAnchor evolves 3 anchors per P output layer # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 [-1, 3, C3, [768]], [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 [-1, 3, C3, [1024]], [-1, 1, Conv, [1280, 3, 2]], # 11-P7/128 [-1, 3, C3, [1280]], [-1, 1, SPPF, [1280, 5]], # 13 ] # YOLOv5 v6.0 head with (P3, P4, P5, P6, P7) outputs head: [[-1, 1, Conv, [1024, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 10], 1, Concat, [1]], # cat backbone P6 [-1, 3, C3, [1024, False]], # 17 [-1, 1, Conv, [768, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P5 [-1, 3, C3, [768, False]], # 21 [-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 25 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 29 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 26], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 32 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 22], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [768, False]], # 35 (P5/32-large) [-1, 1, Conv, [768, 3, 2]], [[-1, 18], 1, Concat, [1]], # cat head P6 [-1, 3, C3, [1024, False]], # 38 (P6/64-xlarge) [-1, 1, Conv, [1024, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P7 [-1, 3, C3, [1280, False]], # 41 (P7/128-xxlarge) [[29, 32, 35, 38, 41], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6, P7) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5-panet.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 PANet head head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 20 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 23 (P5/32-large) [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5l6.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: - [19,27, 44,40, 38,94] # P3/8 - [96,68, 86,152, 180,137] # P4/16 - [140,301, 303,264, 238,542] # P5/32 - [436,615, 739,380, 925,792] # P6/64 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 [-1, 3, C3, [768]], [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 11 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [768, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P5 [-1, 3, C3, [768, False]], # 15 [-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 19 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 23 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 20], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 26 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 16], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [768, False]], # 29 (P5/32-large) [-1, 1, Conv, [768, 3, 2]], [[-1, 12], 1, Concat, [1]], # cat head P6 [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5m6.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 0.67 # model depth multiple width_multiple: 0.75 # layer channel multiple anchors: - [19,27, 44,40, 38,94] # P3/8 - [96,68, 86,152, 180,137] # P4/16 - [140,301, 303,264, 238,542] # P5/32 - [436,615, 739,380, 925,792] # P6/64 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 [-1, 3, C3, [768]], [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 11 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [768, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P5 [-1, 3, C3, [768, False]], # 15 [-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 19 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 23 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 20], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 26 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 16], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [768, False]], # 29 (P5/32-large) [-1, 1, Conv, [768, 3, 2]], [[-1, 12], 1, Concat, [1]], # cat head P6 [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5n6.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 0.33 # model depth multiple width_multiple: 0.25 # layer channel multiple anchors: - [19,27, 44,40, 38,94] # P3/8 - [96,68, 86,152, 180,137] # P4/16 - [140,301, 303,264, 238,542] # P5/32 - [436,615, 739,380, 925,792] # P6/64 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 [-1, 3, C3, [768]], [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 11 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [768, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P5 [-1, 3, C3, [768, False]], # 15 [-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 19 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 23 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 20], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 26 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 16], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [768, False]], # 29 (P5/32-large) [-1, 1, Conv, [768, 3, 2]], [[-1, 12], 1, Concat, [1]], # cat head P6 [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5s-ghost.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 0.33 # model depth multiple width_multiple: 0.50 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, GhostConv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3Ghost, [128]], [-1, 1, GhostConv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3Ghost, [256]], [-1, 1, GhostConv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3Ghost, [512]], [-1, 1, GhostConv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3Ghost, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 head head: [[-1, 1, GhostConv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3Ghost, [512, False]], # 13 [-1, 1, GhostConv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3Ghost, [256, False]], # 17 (P3/8-small) [-1, 1, GhostConv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3Ghost, [512, False]], # 20 (P4/16-medium) [-1, 1, GhostConv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3Ghost, [1024, False]], # 23 (P5/32-large) [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5s-transformer.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 0.33 # model depth multiple width_multiple: 0.50 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3TR, [1024]], # 9 <--- C3TR() Transformer module [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 20 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 23 (P5/32-large) [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5s6.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 0.33 # model depth multiple width_multiple: 0.50 # layer channel multiple anchors: - [19,27, 44,40, 38,94] # P3/8 - [96,68, 86,152, 180,137] # P4/16 - [140,301, 303,264, 238,542] # P5/32 - [436,615, 739,380, 925,792] # P6/64 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 [-1, 3, C3, [768]], [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 11 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [768, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P5 [-1, 3, C3, [768, False]], # 15 [-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 19 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 23 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 20], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 26 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 16], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [768, False]], # 29 (P5/32-large) [-1, 1, Conv, [768, 3, 2]], [[-1, 12], 1, Concat, [1]], # cat head P6 [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) ] ================================================ FILE: RStask/ObjectDetection/models/hub/yolov5x6.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.33 # model depth multiple width_multiple: 1.25 # layer channel multiple anchors: - [19,27, 44,40, 38,94] # P3/8 - [96,68, 86,152, 180,137] # P4/16 - [140,301, 303,264, 238,542] # P5/32 - [436,615, 739,380, 925,792] # P6/64 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 [-1, 3, C3, [768]], [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 11 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [768, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 8], 1, Concat, [1]], # cat backbone P5 [-1, 3, C3, [768, False]], # 15 [-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 19 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 23 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 20], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 26 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 16], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [768, False]], # 29 (P5/32-large) [-1, 1, Conv, [768, 3, 2]], [[-1, 12], 1, Concat, [1]], # cat head P6 [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) ] ================================================ FILE: RStask/ObjectDetection/models/tf.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ TensorFlow, Keras and TFLite versions of YOLOv5 Authored by https://github.com/zldrobit in PR https://github.com/ultralytics/yolov5/pull/1127 Usage: $ python models/tf.py --weights yolov5s.pt Export: $ python path/to/export.py --weights yolov5s.pt --include saved_model pb tflite tfjs """ import argparse import sys from copy import deepcopy from pathlib import Path FILE = Path(__file__).resolve() ROOT = FILE.parents[1] # YOLOv5 root directory if str(ROOT) not in sys.path: sys.path.append(str(ROOT)) # add ROOT to PATH # ROOT = ROOT.relative_to(Path.cwd()) # relative import numpy as np import tensorflow as tf import torch import torch.nn as nn from tensorflow import keras from models.common import C3, SPP, SPPF, Bottleneck, BottleneckCSP, Concat, Conv, DWConv, Focus, autopad from models.experimental import CrossConv, MixConv2d, attempt_load from models.yolo import Detect from utils.activations import SiLU from utils.general import LOGGER, make_divisible, print_args class TFBN(keras.layers.Layer): # TensorFlow BatchNormalization wrapper def __init__(self, w=None): super().__init__() self.bn = keras.layers.BatchNormalization( beta_initializer=keras.initializers.Constant(w.bias.numpy()), gamma_initializer=keras.initializers.Constant(w.weight.numpy()), moving_mean_initializer=keras.initializers.Constant(w.running_mean.numpy()), moving_variance_initializer=keras.initializers.Constant(w.running_var.numpy()), epsilon=w.eps) def call(self, inputs): return self.bn(inputs) class TFPad(keras.layers.Layer): def __init__(self, pad): super().__init__() self.pad = tf.constant([[0, 0], [pad, pad], [pad, pad], [0, 0]]) def call(self, inputs): return tf.pad(inputs, self.pad, mode='constant', constant_values=0) class TFConv(keras.layers.Layer): # Standard convolution def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True, w=None): # ch_in, ch_out, weights, kernel, stride, padding, groups super().__init__() assert g == 1, "TF v2.2 Conv2D does not support 'groups' argument" assert isinstance(k, int), "Convolution with multiple kernels are not allowed." # TensorFlow convolution padding is inconsistent with PyTorch (e.g. k=3 s=2 'SAME' padding) # see https://stackoverflow.com/questions/52975843/comparing-conv2d-with-padding-between-tensorflow-and-pytorch conv = keras.layers.Conv2D( c2, k, s, 'SAME' if s == 1 else 'VALID', use_bias=False if hasattr(w, 'bn') else True, kernel_initializer=keras.initializers.Constant(w.conv.weight.permute(2, 3, 1, 0).numpy()), bias_initializer='zeros' if hasattr(w, 'bn') else keras.initializers.Constant(w.conv.bias.numpy())) self.conv = conv if s == 1 else keras.Sequential([TFPad(autopad(k, p)), conv]) self.bn = TFBN(w.bn) if hasattr(w, 'bn') else tf.identity # YOLOv5 activations if isinstance(w.act, nn.LeakyReLU): self.act = (lambda x: keras.activations.relu(x, alpha=0.1)) if act else tf.identity elif isinstance(w.act, nn.Hardswish): self.act = (lambda x: x * tf.nn.relu6(x + 3) * 0.166666667) if act else tf.identity elif isinstance(w.act, (nn.SiLU, SiLU)): self.act = (lambda x: keras.activations.swish(x)) if act else tf.identity else: raise Exception(f'no matching TensorFlow activation found for {w.act}') def call(self, inputs): return self.act(self.bn(self.conv(inputs))) class TFFocus(keras.layers.Layer): # Focus wh information into c-space def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True, w=None): # ch_in, ch_out, kernel, stride, padding, groups super().__init__() self.conv = TFConv(c1 * 4, c2, k, s, p, g, act, w.conv) def call(self, inputs): # x(b,w,h,c) -> y(b,w/2,h/2,4c) # inputs = inputs / 255 # normalize 0-255 to 0-1 return self.conv( tf.concat( [inputs[:, ::2, ::2, :], inputs[:, 1::2, ::2, :], inputs[:, ::2, 1::2, :], inputs[:, 1::2, 1::2, :]], 3)) class TFBottleneck(keras.layers.Layer): # Standard bottleneck def __init__(self, c1, c2, shortcut=True, g=1, e=0.5, w=None): # ch_in, ch_out, shortcut, groups, expansion super().__init__() c_ = int(c2 * e) # hidden channels self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) self.cv2 = TFConv(c_, c2, 3, 1, g=g, w=w.cv2) self.add = shortcut and c1 == c2 def call(self, inputs): return inputs + self.cv2(self.cv1(inputs)) if self.add else self.cv2(self.cv1(inputs)) class TFConv2d(keras.layers.Layer): # Substitution for PyTorch nn.Conv2D def __init__(self, c1, c2, k, s=1, g=1, bias=True, w=None): super().__init__() assert g == 1, "TF v2.2 Conv2D does not support 'groups' argument" self.conv = keras.layers.Conv2D( c2, k, s, 'VALID', use_bias=bias, kernel_initializer=keras.initializers.Constant(w.weight.permute(2, 3, 1, 0).numpy()), bias_initializer=keras.initializers.Constant(w.bias.numpy()) if bias else None, ) def call(self, inputs): return self.conv(inputs) class TFBottleneckCSP(keras.layers.Layer): # CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5, w=None): # ch_in, ch_out, number, shortcut, groups, expansion super().__init__() c_ = int(c2 * e) # hidden channels self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) self.cv2 = TFConv2d(c1, c_, 1, 1, bias=False, w=w.cv2) self.cv3 = TFConv2d(c_, c_, 1, 1, bias=False, w=w.cv3) self.cv4 = TFConv(2 * c_, c2, 1, 1, w=w.cv4) self.bn = TFBN(w.bn) self.act = lambda x: keras.activations.swish(x) self.m = keras.Sequential([TFBottleneck(c_, c_, shortcut, g, e=1.0, w=w.m[j]) for j in range(n)]) def call(self, inputs): y1 = self.cv3(self.m(self.cv1(inputs))) y2 = self.cv2(inputs) return self.cv4(self.act(self.bn(tf.concat((y1, y2), axis=3)))) class TFC3(keras.layers.Layer): # CSP Bottleneck with 3 convolutions def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5, w=None): # ch_in, ch_out, number, shortcut, groups, expansion super().__init__() c_ = int(c2 * e) # hidden channels self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) self.cv2 = TFConv(c1, c_, 1, 1, w=w.cv2) self.cv3 = TFConv(2 * c_, c2, 1, 1, w=w.cv3) self.m = keras.Sequential([TFBottleneck(c_, c_, shortcut, g, e=1.0, w=w.m[j]) for j in range(n)]) def call(self, inputs): return self.cv3(tf.concat((self.m(self.cv1(inputs)), self.cv2(inputs)), axis=3)) class TFSPP(keras.layers.Layer): # Spatial pyramid pooling layer used in YOLOv3-SPP def __init__(self, c1, c2, k=(5, 9, 13), w=None): super().__init__() c_ = c1 // 2 # hidden channels self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) self.cv2 = TFConv(c_ * (len(k) + 1), c2, 1, 1, w=w.cv2) self.m = [keras.layers.MaxPool2D(pool_size=x, strides=1, padding='SAME') for x in k] def call(self, inputs): x = self.cv1(inputs) return self.cv2(tf.concat([x] + [m(x) for m in self.m], 3)) class TFSPPF(keras.layers.Layer): # Spatial pyramid pooling-Fast layer def __init__(self, c1, c2, k=5, w=None): super().__init__() c_ = c1 // 2 # hidden channels self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) self.cv2 = TFConv(c_ * 4, c2, 1, 1, w=w.cv2) self.m = keras.layers.MaxPool2D(pool_size=k, strides=1, padding='SAME') def call(self, inputs): x = self.cv1(inputs) y1 = self.m(x) y2 = self.m(y1) return self.cv2(tf.concat([x, y1, y2, self.m(y2)], 3)) class TFDetect(keras.layers.Layer): # TF YOLOv5 Detect layer def __init__(self, nc=80, anchors=(), ch=(), imgsz=(640, 640), w=None): # detection layer super().__init__() self.stride = tf.convert_to_tensor(w.stride.numpy(), dtype=tf.float32) self.nc = nc # number of classes self.no = nc + 5 # number of outputs per anchor self.nl = len(anchors) # number of detection layers self.na = len(anchors[0]) // 2 # number of anchors self.grid = [tf.zeros(1)] * self.nl # init grid self.anchors = tf.convert_to_tensor(w.anchors.numpy(), dtype=tf.float32) self.anchor_grid = tf.reshape(self.anchors * tf.reshape(self.stride, [self.nl, 1, 1]), [self.nl, 1, -1, 1, 2]) self.m = [TFConv2d(x, self.no * self.na, 1, w=w.m[i]) for i, x in enumerate(ch)] self.training = False # set to False after building model self.imgsz = imgsz for i in range(self.nl): ny, nx = self.imgsz[0] // self.stride[i], self.imgsz[1] // self.stride[i] self.grid[i] = self._make_grid(nx, ny) def call(self, inputs): z = [] # inference output x = [] for i in range(self.nl): x.append(self.m[i](inputs[i])) # x(bs,20,20,255) to x(bs,3,20,20,85) ny, nx = self.imgsz[0] // self.stride[i], self.imgsz[1] // self.stride[i] x[i] = tf.reshape(x[i], [-1, ny * nx, self.na, self.no]) if not self.training: # inference y = tf.sigmoid(x[i]) grid = tf.transpose(self.grid[i], [0, 2, 1, 3]) - 0.5 anchor_grid = tf.transpose(self.anchor_grid[i], [0, 2, 1, 3]) * 4 xy = (y[..., 0:2] * 2 + grid) * self.stride[i] # xy wh = y[..., 2:4] ** 2 * anchor_grid # Normalize xywh to 0-1 to reduce calibration error xy /= tf.constant([[self.imgsz[1], self.imgsz[0]]], dtype=tf.float32) wh /= tf.constant([[self.imgsz[1], self.imgsz[0]]], dtype=tf.float32) y = tf.concat([xy, wh, y[..., 4:]], -1) z.append(tf.reshape(y, [-1, self.na * ny * nx, self.no])) return tf.transpose(x, [0, 2, 1, 3]) if self.training else (tf.concat(z, 1), x) @staticmethod def _make_grid(nx=20, ny=20): # yv, xv = torch.meshgrid([torch.arange(ny), torch.arange(nx)]) # return torch.stack((xv, yv), 2).view((1, 1, ny, nx, 2)).float() xv, yv = tf.meshgrid(tf.range(nx), tf.range(ny)) return tf.cast(tf.reshape(tf.stack([xv, yv], 2), [1, 1, ny * nx, 2]), dtype=tf.float32) class TFUpsample(keras.layers.Layer): # TF version of torch.nn.Upsample() def __init__(self, size, scale_factor, mode, w=None): # warning: all arguments needed including 'w' super().__init__() assert scale_factor == 2, "scale_factor must be 2" self.upsample = lambda x: tf.image.resize(x, (x.shape[1] * 2, x.shape[2] * 2), method=mode) # self.upsample = keras.layers.UpSampling2D(size=scale_factor, interpolation=mode) # with default arguments: align_corners=False, half_pixel_centers=False # self.upsample = lambda x: tf.raw_ops.ResizeNearestNeighbor(images=x, # size=(x.shape[1] * 2, x.shape[2] * 2)) def call(self, inputs): return self.upsample(inputs) class TFConcat(keras.layers.Layer): # TF version of torch.concat() def __init__(self, dimension=1, w=None): super().__init__() assert dimension == 1, "convert only NCHW to NHWC concat" self.d = 3 def call(self, inputs): return tf.concat(inputs, self.d) def parse_model(d, ch, model, imgsz): # model_dict, input_channels(3) LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10} {'module':<40}{'arguments':<30}") anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'] na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors no = na * (nc + 5) # number of outputs = anchors * (classes + 5) layers, save, c2 = [], [], ch[-1] # layers, savelist, ch out for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): # from, number, module, args m_str = m m = eval(m) if isinstance(m, str) else m # eval strings for j, a in enumerate(args): try: args[j] = eval(a) if isinstance(a, str) else a # eval strings except NameError: pass n = max(round(n * gd), 1) if n > 1 else n # depth gain if m in [nn.Conv2d, Conv, Bottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3]: c1, c2 = ch[f], args[0] c2 = make_divisible(c2 * gw, 8) if c2 != no else c2 args = [c1, c2, *args[1:]] if m in [BottleneckCSP, C3]: args.insert(2, n) n = 1 elif m is nn.BatchNorm2d: args = [ch[f]] elif m is Concat: c2 = sum(ch[-1 if x == -1 else x + 1] for x in f) elif m is Detect: args.append([ch[x + 1] for x in f]) if isinstance(args[1], int): # number of anchors args[1] = [list(range(args[1] * 2))] * len(f) args.append(imgsz) else: c2 = ch[f] tf_m = eval('TF' + m_str.replace('nn.', '')) m_ = keras.Sequential([tf_m(*args, w=model.model[i][j]) for j in range(n)]) if n > 1 \ else tf_m(*args, w=model.model[i]) # module torch_m_ = nn.Sequential(*(m(*args) for _ in range(n))) if n > 1 else m(*args) # module t = str(m)[8:-2].replace('__main__.', '') # module type np = sum(x.numel() for x in torch_m_.parameters()) # number params m_.i, m_.f, m_.type, m_.np = i, f, t, np # attach index, 'from' index, type, number params LOGGER.info(f'{i:>3}{str(f):>18}{str(n):>3}{np:>10} {t:<40}{str(args):<30}') # print save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1) # append to savelist layers.append(m_) ch.append(c2) return keras.Sequential(layers), sorted(save) class TFModel: # TF YOLOv5 model def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, model=None, imgsz=(640, 640)): # model, channels, classes super().__init__() if isinstance(cfg, dict): self.yaml = cfg # model dict else: # is *.yaml import yaml # for torch hub self.yaml_file = Path(cfg).name with open(cfg) as f: self.yaml = yaml.load(f, Loader=yaml.FullLoader) # model dict # Define model if nc and nc != self.yaml['nc']: LOGGER.info(f"Overriding {cfg} nc={self.yaml['nc']} with nc={nc}") self.yaml['nc'] = nc # override yaml value self.model, self.savelist = parse_model(deepcopy(self.yaml), ch=[ch], model=model, imgsz=imgsz) def predict(self, inputs, tf_nms=False, agnostic_nms=False, topk_per_class=100, topk_all=100, iou_thres=0.45, conf_thres=0.25): y = [] # outputs x = inputs for i, m in enumerate(self.model.layers): if m.f != -1: # if not from previous layer x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers x = m(x) # run y.append(x if m.i in self.savelist else None) # save output # Add TensorFlow NMS if tf_nms: boxes = self._xywh2xyxy(x[0][..., :4]) probs = x[0][:, :, 4:5] classes = x[0][:, :, 5:] scores = probs * classes if agnostic_nms: nms = AgnosticNMS()((boxes, classes, scores), topk_all, iou_thres, conf_thres) return nms, x[1] else: boxes = tf.expand_dims(boxes, 2) nms = tf.image.combined_non_max_suppression(boxes, scores, topk_per_class, topk_all, iou_thres, conf_thres, clip_boxes=False) return nms, x[1] return x[0] # output only first tensor [1,6300,85] = [xywh, conf, class0, class1, ...] # x = x[0][0] # [x(1,6300,85), ...] to x(6300,85) # xywh = x[..., :4] # x(6300,4) boxes # conf = x[..., 4:5] # x(6300,1) confidences # cls = tf.reshape(tf.cast(tf.argmax(x[..., 5:], axis=1), tf.float32), (-1, 1)) # x(6300,1) classes # return tf.concat([conf, cls, xywh], 1) @staticmethod def _xywh2xyxy(xywh): # Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right x, y, w, h = tf.split(xywh, num_or_size_splits=4, axis=-1) return tf.concat([x - w / 2, y - h / 2, x + w / 2, y + h / 2], axis=-1) class AgnosticNMS(keras.layers.Layer): # TF Agnostic NMS def call(self, input, topk_all, iou_thres, conf_thres): # wrap map_fn to avoid TypeSpec related error https://stackoverflow.com/a/65809989/3036450 return tf.map_fn(lambda x: self._nms(x, topk_all, iou_thres, conf_thres), input, fn_output_signature=(tf.float32, tf.float32, tf.float32, tf.int32), name='agnostic_nms') @staticmethod def _nms(x, topk_all=100, iou_thres=0.45, conf_thres=0.25): # agnostic NMS boxes, classes, scores = x class_inds = tf.cast(tf.argmax(classes, axis=-1), tf.float32) scores_inp = tf.reduce_max(scores, -1) selected_inds = tf.image.non_max_suppression(boxes, scores_inp, max_output_size=topk_all, iou_threshold=iou_thres, score_threshold=conf_thres) selected_boxes = tf.gather(boxes, selected_inds) padded_boxes = tf.pad(selected_boxes, paddings=[[0, topk_all - tf.shape(selected_boxes)[0]], [0, 0]], mode="CONSTANT", constant_values=0.0) selected_scores = tf.gather(scores_inp, selected_inds) padded_scores = tf.pad(selected_scores, paddings=[[0, topk_all - tf.shape(selected_boxes)[0]]], mode="CONSTANT", constant_values=-1.0) selected_classes = tf.gather(class_inds, selected_inds) padded_classes = tf.pad(selected_classes, paddings=[[0, topk_all - tf.shape(selected_boxes)[0]]], mode="CONSTANT", constant_values=-1.0) valid_detections = tf.shape(selected_inds)[0] return padded_boxes, padded_scores, padded_classes, valid_detections def representative_dataset_gen(dataset, ncalib=100): # Representative dataset generator for use with converter.representative_dataset, returns a generator of np arrays for n, (path, img, im0s, vid_cap, string) in enumerate(dataset): input = np.transpose(img, [1, 2, 0]) input = np.expand_dims(input, axis=0).astype(np.float32) input /= 255 yield [input] if n >= ncalib: break def run( weights=ROOT / 'yolov5s.pt', # weights path imgsz=(640, 640), # inference size h,w batch_size=1, # batch size dynamic=False, # dynamic batch size ): # PyTorch model im = torch.zeros((batch_size, 3, *imgsz)) # BCHW image model = attempt_load(weights, map_location=torch.device('cpu'), inplace=True, fuse=False) _ = model(im) # inference model.info() # TensorFlow model im = tf.zeros((batch_size, *imgsz, 3)) # BHWC image tf_model = TFModel(cfg=model.yaml, model=model, nc=model.nc, imgsz=imgsz) _ = tf_model.predict(im) # inference # Keras model im = keras.Input(shape=(*imgsz, 3), batch_size=None if dynamic else batch_size) keras_model = keras.Model(inputs=im, outputs=tf_model.predict(im)) keras_model.summary() LOGGER.info('PyTorch, TensorFlow and Keras models successfully verified.\nUse export.py for TF model export.') def parse_opt(): parser = argparse.ArgumentParser() parser.add_argument('--weights', type=str, default=ROOT / 'yolov5s.pt', help='weights path') parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640], help='inference size h,w') parser.add_argument('--batch-size', type=int, default=1, help='batch size') parser.add_argument('--dynamic', action='store_true', help='dynamic batch size') opt = parser.parse_args() opt.imgsz *= 2 if len(opt.imgsz) == 1 else 1 # expand print_args(vars(opt)) return opt def main(opt): run(**vars(opt)) if __name__ == "__main__": opt = parse_opt() main(opt) ================================================ FILE: RStask/ObjectDetection/models/yolo.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ YOLO-specific modules Usage: $ python path/to/models/yolo.py --cfg yolov5s.yaml """ import argparse import os import platform import sys from copy import deepcopy from pathlib import Path FILE = Path(__file__).resolve() ROOT = FILE.parents[1] # YOLOv5 root directory if str(ROOT) not in sys.path: sys.path.append(str(ROOT)) # add ROOT to PATH if platform.system() != 'Windows': ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative from models.common import * from models.experimental import * from utils.autoanchor import check_anchor_order from utils.general import LOGGER, check_version, check_yaml, make_divisible, print_args from utils.plots import feature_visualization from utils.torch_utils import (fuse_conv_and_bn, initialize_weights, model_info, profile, scale_img, select_device, time_sync) try: import thop # for FLOPs computation except ImportError: thop = None class Detect(nn.Module): stride = None # strides computed during build onnx_dynamic = False # ONNX export parameter export = False # export mode def __init__(self, nc=80, anchors=(), ch=(), inplace=True): # detection layer super().__init__() self.nc = nc # number of classes self.no = nc + 5 # number of outputs per anchor self.nl = len(anchors) # number of detection layers self.na = len(anchors[0]) // 2 # number of anchors self.grid = [torch.zeros(1)] * self.nl # init grid self.anchor_grid = [torch.zeros(1)] * self.nl # init anchor grid self.register_buffer('anchors', torch.tensor(anchors).float().view(self.nl, -1, 2)) # shape(nl,na,2) self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv self.inplace = inplace # use in-place ops (e.g. slice assignment) def forward(self, x): z = [] # inference output for i in range(self.nl): x[i] = self.m[i](x[i]) # conv bs, _, ny, nx = x[i].shape # x(bs,255,20,20) to x(bs,3,20,20,85) x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous() if not self.training: # inference if self.onnx_dynamic or self.grid[i].shape[2:4] != x[i].shape[2:4]: self.grid[i], self.anchor_grid[i] = self._make_grid(nx, ny, i) y = x[i].sigmoid() if self.inplace: y[..., 0:2] = (y[..., 0:2] * 2 + self.grid[i]) * self.stride[i] # xy y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh else: # for YOLOv5 on AWS Inferentia https://github.com/ultralytics/yolov5/pull/2953 xy, wh, conf = y.split((2, 2, self.nc + 1), 4) # y.tensor_split((2, 4, 5), 4) # torch 1.8.0 xy = (xy * 2 + self.grid[i]) * self.stride[i] # xy wh = (wh * 2) ** 2 * self.anchor_grid[i] # wh y = torch.cat((xy, wh, conf), 4) z.append(y.view(bs, -1, self.no)) return x if self.training else (torch.cat(z, 1),) if self.export else (torch.cat(z, 1), x) def _make_grid(self, nx=20, ny=20, i=0): d = self.anchors[i].device t = self.anchors[i].dtype shape = 1, self.na, ny, nx, 2 # grid shape y, x = torch.arange(ny, device=d, dtype=t), torch.arange(nx, device=d, dtype=t) if check_version(torch.__version__, '1.10.0'): # torch>=1.10.0 meshgrid workaround for torch>=0.7 compatibility yv, xv = torch.meshgrid(y, x, indexing='ij') else: yv, xv = torch.meshgrid(y, x) grid = torch.stack((xv, yv), 2).expand(shape) - 0.5 # add grid offset, i.e. y = 2.0 * x - 0.5 anchor_grid = (self.anchors[i] * self.stride[i]).view((1, self.na, 1, 1, 2)).expand(shape) return grid, anchor_grid class Model(nn.Module): # YOLOv5 model def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, anchors=None): # model, input channels, number of classes super().__init__() if isinstance(cfg, dict): self.yaml = cfg # model dict else: # is *.yaml import yaml # for torch hub self.yaml_file = Path(cfg).name with open(cfg, encoding='ascii', errors='ignore') as f: self.yaml = yaml.safe_load(f) # model dict # Define model ch = self.yaml['ch'] = self.yaml.get('ch', ch) # input channels if nc and nc != self.yaml['nc']: LOGGER.info(f"Overriding model.yaml nc={self.yaml['nc']} with nc={nc}") self.yaml['nc'] = nc # override yaml value if anchors: LOGGER.info(f'Overriding model.yaml anchors with anchors={anchors}') self.yaml['anchors'] = round(anchors) # override yaml value self.model, self.save = parse_model(deepcopy(self.yaml), ch=[ch]) # model, savelist self.names = [str(i) for i in range(self.yaml['nc'])] # default names self.inplace = self.yaml.get('inplace', True) # Build strides, anchors m = self.model[-1] # Detect() if isinstance(m, Detect): s = 256 # 2x min stride m.inplace = self.inplace m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))]) # forward check_anchor_order(m) # must be in pixel-space (not grid-space) m.anchors /= m.stride.view(-1, 1, 1) self.stride = m.stride self._initialize_biases() # only run once # Init weights, biases initialize_weights(self) self.info() LOGGER.info('') def forward(self, x, augment=False, profile=False, visualize=False): if augment: return self._forward_augment(x) # augmented inference, None return self._forward_once(x, profile, visualize) # single-scale inference, train def _forward_augment(self, x): img_size = x.shape[-2:] # height, width s = [1, 0.83, 0.67] # scales f = [None, 3, None] # flips (2-ud, 3-lr) y = [] # outputs for si, fi in zip(s, f): xi = scale_img(x.flip(fi) if fi else x, si, gs=int(self.stride.max())) yi = self._forward_once(xi)[0] # forward # cv2.imwrite(f'img_{si}.jpg', 255 * xi[0].cpu().numpy().transpose((1, 2, 0))[:, :, ::-1]) # save yi = self._descale_pred(yi, fi, si, img_size) y.append(yi) y = self._clip_augmented(y) # clip augmented tails return torch.cat(y, 1), None # augmented inference, train def _forward_once(self, x, profile=False, visualize=False): y, dt = [], [] # outputs for m in self.model: if m.f != -1: # if not from previous layer x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers if profile: self._profile_one_layer(m, x, dt) x = m(x) # run y.append(x if m.i in self.save else None) # save output if visualize: feature_visualization(x, m.type, m.i, save_dir=visualize) return x def _descale_pred(self, p, flips, scale, img_size): # de-scale predictions following augmented inference (inverse operation) if self.inplace: p[..., :4] /= scale # de-scale if flips == 2: p[..., 1] = img_size[0] - p[..., 1] # de-flip ud elif flips == 3: p[..., 0] = img_size[1] - p[..., 0] # de-flip lr else: x, y, wh = p[..., 0:1] / scale, p[..., 1:2] / scale, p[..., 2:4] / scale # de-scale if flips == 2: y = img_size[0] - y # de-flip ud elif flips == 3: x = img_size[1] - x # de-flip lr p = torch.cat((x, y, wh, p[..., 4:]), -1) return p def _clip_augmented(self, y): # Clip YOLOv5 augmented inference tails nl = self.model[-1].nl # number of detection layers (P3-P5) g = sum(4 ** x for x in range(nl)) # grid points e = 1 # exclude layer count i = (y[0].shape[1] // g) * sum(4 ** x for x in range(e)) # indices y[0] = y[0][:, :-i] # large i = (y[-1].shape[1] // g) * sum(4 ** (nl - 1 - x) for x in range(e)) # indices y[-1] = y[-1][:, i:] # small return y def _profile_one_layer(self, m, x, dt): c = isinstance(m, Detect) # is final layer, copy input as inplace fix o = thop.profile(m, inputs=(x.copy() if c else x,), verbose=False)[0] / 1E9 * 2 if thop else 0 # FLOPs t = time_sync() for _ in range(10): m(x.copy() if c else x) dt.append((time_sync() - t) * 100) if m == self.model[0]: LOGGER.info(f"{'time (ms)':>10s} {'GFLOPs':>10s} {'params':>10s} {'module'}") LOGGER.info(f'{dt[-1]:10.2f} {o:10.2f} {m.np:10.0f} {m.type}') if c: LOGGER.info(f"{sum(dt):10.2f} {'-':>10s} {'-':>10s} Total") def _initialize_biases(self, cf=None): # initialize biases into Detect(), cf is class frequency # https://arxiv.org/abs/1708.02002 section 3.3 # cf = torch.bincount(torch.tensor(np.concatenate(dataset.labels, 0)[:, 0]).long(), minlength=nc) + 1. m = self.model[-1] # Detect() module for mi, s in zip(m.m, m.stride): # from b = mi.bias.view(m.na, -1) # conv.bias(255) to (3,85) b.data[:, 4] += math.log(8 / (640 / s) ** 2) # obj (8 objects per 640 image) b.data[:, 5:] += math.log(0.6 / (m.nc - 0.999999)) if cf is None else torch.log(cf / cf.sum()) # cls mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True) def _print_biases(self): m = self.model[-1] # Detect() module for mi in m.m: # from b = mi.bias.detach().view(m.na, -1).T # conv.bias(255) to (3,85) LOGGER.info( ('%6g Conv2d.bias:' + '%10.3g' * 6) % (mi.weight.shape[1], *b[:5].mean(1).tolist(), b[5:].mean())) # def _print_weights(self): # for m in self.model.modules(): # if type(m) is Bottleneck: # LOGGER.info('%10.3g' % (m.w.detach().sigmoid() * 2)) # shortcut weights def fuse(self): # fuse model Conv2d() + BatchNorm2d() layers for m in self.model.modules(): if isinstance(m, (Conv, DWConv)) and hasattr(m, 'bn'): m.conv = fuse_conv_and_bn(m.conv, m.bn) # update conv delattr(m, 'bn') # remove batchnorm m.forward = m.forward_fuse # update forward self.info() return self def info(self, verbose=False, img_size=640): # print model information model_info(self, verbose, img_size) def _apply(self, fn): # Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers self = super()._apply(fn) m = self.model[-1] # Detect() if isinstance(m, Detect): m.stride = fn(m.stride) m.grid = list(map(fn, m.grid)) if isinstance(m.anchor_grid, list): m.anchor_grid = list(map(fn, m.anchor_grid)) return self def parse_model(d, ch): # model_dict, input_channels(3) LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10} {'module':<40}{'arguments':<30}") anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'] na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors no = na * (nc + 5) # number of outputs = anchors * (classes + 5) layers, save, c2 = [], [], ch[-1] # layers, savelist, ch out for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): # from, number, module, args m = eval(m) if isinstance(m, str) else m # eval strings for j, a in enumerate(args): try: args[j] = eval(a) if isinstance(a, str) else a # eval strings except NameError: pass n = n_ = max(round(n * gd), 1) if n > 1 else n # depth gain if m in (Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3, C3TR, C3SPP, C3Ghost): c1, c2 = ch[f], args[0] if c2 != no: # if not output c2 = make_divisible(c2 * gw, 8) args = [c1, c2, *args[1:]] if m in [BottleneckCSP, C3, C3TR, C3Ghost]: args.insert(2, n) # number of repeats n = 1 elif m is nn.BatchNorm2d: args = [ch[f]] elif m is Concat: c2 = sum(ch[x] for x in f) elif m is Detect: args.append([ch[x] for x in f]) if isinstance(args[1], int): # number of anchors args[1] = [list(range(args[1] * 2))] * len(f) elif m is Contract: c2 = ch[f] * args[0] ** 2 elif m is Expand: c2 = ch[f] // args[0] ** 2 else: c2 = ch[f] m_ = nn.Sequential(*(m(*args) for _ in range(n))) if n > 1 else m(*args) # module t = str(m)[8:-2].replace('__main__.', '') # module type np = sum(x.numel() for x in m_.parameters()) # number params m_.i, m_.f, m_.type, m_.np = i, f, t, np # attach index, 'from' index, type, number params LOGGER.info(f'{i:>3}{str(f):>18}{n_:>3}{np:10.0f} {t:<40}{str(args):<30}') # print save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1) # append to savelist layers.append(m_) if i == 0: ch = [] ch.append(c2) return nn.Sequential(*layers), sorted(save) if __name__ == '__main__': parser = argparse.ArgumentParser() parser.add_argument('--cfg', type=str, default='yolov5s.yaml', help='model.yaml') parser.add_argument('--batch-size', type=int, default=1, help='total batch size for all GPUs') parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') parser.add_argument('--profile', action='store_true', help='profile model speed') parser.add_argument('--line-profile', action='store_true', help='profile model speed layer by layer') parser.add_argument('--test', action='store_true', help='test all yolo*.yaml') opt = parser.parse_args() opt.cfg = check_yaml(opt.cfg) # check YAML print_args(vars(opt)) device = select_device(opt.device) # Create model im = torch.rand(opt.batch_size, 3, 640, 640).to(device) model = Model(opt.cfg).to(device) # Options if opt.line_profile: # profile layer by layer _ = model(im, profile=True) elif opt.profile: # profile forward-backward results = profile(input=im, ops=[model], n=3) elif opt.test: # test all models for cfg in Path(ROOT / 'models').rglob('yolo*.yaml'): try: _ = Model(cfg) except Exception as e: print(f'Error in {cfg}: {e}') ================================================ FILE: RStask/ObjectDetection/models/yolov5l.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.0 # model depth multiple width_multiple: 1.0 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 20 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 23 (P5/32-large) [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/yolov5m.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 0.67 # model depth multiple width_multiple: 0.75 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 20 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 23 (P5/32-large) [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/yolov5n.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 0.33 # model depth multiple width_multiple: 0.25 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 20 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 23 (P5/32-large) [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/yolov5s.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 0.33 # model depth multiple width_multiple: 0.50 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 20 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 23 (P5/32-large) [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/models/yolov5x.yaml ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Parameters nc: 80 # number of classes depth_multiple: 1.33 # model depth multiple width_multiple: 1.25 # layer channel multiple anchors: - [10,13, 16,30, 33,23] # P3/8 - [30,61, 62,45, 59,119] # P4/16 - [116,90, 156,198, 373,326] # P5/32 # YOLOv5 v6.0 backbone backbone: # [from, number, module, args] [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3, C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 6, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]], # 9 ] # YOLOv5 v6.0 head head: [[-1, 1, Conv, [512, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 6], 1, Concat, [1]], # cat backbone P4 [-1, 3, C3, [512, False]], # 13 [-1, 1, Conv, [256, 1, 1]], [-1, 1, nn.Upsample, [None, 2, 'nearest']], [[-1, 4], 1, Concat, [1]], # cat backbone P3 [-1, 3, C3, [256, False]], # 17 (P3/8-small) [-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]], # cat head P4 [-1, 3, C3, [512, False]], # 20 (P4/16-medium) [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]], # cat head P5 [-1, 3, C3, [1024, False]], # 23 (P5/32-large) [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) ] ================================================ FILE: RStask/ObjectDetection/test.py ================================================ from RStask.ObjectDetection.YOLOv5 import YoloDetection model=YoloDetection('cuda:0') det=model.inference('/data/haonan.guo/RSChatGPT/test.tif',None,'/data/haonan.guo/RSChatGPT/output.png') ================================================ FILE: RStask/ObjectDetection/utils/__init__.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ utils/initialization """ def notebook_init(verbose=True): # Check system software and hardware print('Checking setup...') import os import shutil from utils.general import check_requirements, emojis, is_colab from utils.torch_utils import select_device # imports check_requirements(('psutil', 'IPython')) import psutil from IPython import display # to display images and clear console output if is_colab(): shutil.rmtree('/content/sample_data', ignore_errors=True) # remove colab /sample_data directory # System info if verbose: gb = 1 << 30 # bytes to GiB (1024 ** 3) ram = psutil.virtual_memory().total total, used, free = shutil.disk_usage("/") display.clear_output() s = f'({os.cpu_count()} CPUs, {ram / gb:.1f} GB RAM, {(total - free) / gb:.1f}/{total / gb:.1f} GB disk)' else: s = '' select_device(newline=False) print(emojis(f'Setup complete ✅ {s}')) return display ================================================ FILE: RStask/ObjectDetection/utils/activations.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Activation functions """ import torch import torch.nn as nn import torch.nn.functional as F class SiLU(nn.Module): # SiLU activation https://arxiv.org/pdf/1606.08415.pdf @staticmethod def forward(x): return x * torch.sigmoid(x) class Hardswish(nn.Module): # Hard-SiLU activation @staticmethod def forward(x): # return x * F.hardsigmoid(x) # for TorchScript and CoreML return x * F.hardtanh(x + 3, 0.0, 6.0) / 6.0 # for TorchScript, CoreML and ONNX class Mish(nn.Module): # Mish activation https://github.com/digantamisra98/Mish @staticmethod def forward(x): return x * F.softplus(x).tanh() class MemoryEfficientMish(nn.Module): # Mish activation memory-efficient class F(torch.autograd.Function): @staticmethod def forward(ctx, x): ctx.save_for_backward(x) return x.mul(torch.tanh(F.softplus(x))) # x * tanh(ln(1 + exp(x))) @staticmethod def backward(ctx, grad_output): x = ctx.saved_tensors[0] sx = torch.sigmoid(x) fx = F.softplus(x).tanh() return grad_output * (fx + x * sx * (1 - fx * fx)) def forward(self, x): return self.F.apply(x) class FReLU(nn.Module): # FReLU activation https://arxiv.org/abs/2007.11824 def __init__(self, c1, k=3): # ch_in, kernel super().__init__() self.conv = nn.Conv2d(c1, c1, k, 1, 1, groups=c1, bias=False) self.bn = nn.BatchNorm2d(c1) def forward(self, x): return torch.max(x, self.bn(self.conv(x))) class AconC(nn.Module): r""" ACON activation (activate or not) AconC: (p1*x-p2*x) * sigmoid(beta*(p1*x-p2*x)) + p2*x, beta is a learnable parameter according to "Activate or Not: Learning Customized Activation" . """ def __init__(self, c1): super().__init__() self.p1 = nn.Parameter(torch.randn(1, c1, 1, 1)) self.p2 = nn.Parameter(torch.randn(1, c1, 1, 1)) self.beta = nn.Parameter(torch.ones(1, c1, 1, 1)) def forward(self, x): dpx = (self.p1 - self.p2) * x return dpx * torch.sigmoid(self.beta * dpx) + self.p2 * x class MetaAconC(nn.Module): r""" ACON activation (activate or not) MetaAconC: (p1*x-p2*x) * sigmoid(beta*(p1*x-p2*x)) + p2*x, beta is generated by a small network according to "Activate or Not: Learning Customized Activation" . """ def __init__(self, c1, k=1, s=1, r=16): # ch_in, kernel, stride, r super().__init__() c2 = max(r, c1 // r) self.p1 = nn.Parameter(torch.randn(1, c1, 1, 1)) self.p2 = nn.Parameter(torch.randn(1, c1, 1, 1)) self.fc1 = nn.Conv2d(c1, c2, k, s, bias=True) self.fc2 = nn.Conv2d(c2, c1, k, s, bias=True) # self.bn1 = nn.BatchNorm2d(c2) # self.bn2 = nn.BatchNorm2d(c1) def forward(self, x): y = x.mean(dim=2, keepdims=True).mean(dim=3, keepdims=True) # batch-size 1 bug/instabilities https://github.com/ultralytics/yolov5/issues/2891 # beta = torch.sigmoid(self.bn2(self.fc2(self.bn1(self.fc1(y))))) # bug/unstable beta = torch.sigmoid(self.fc2(self.fc1(y))) # bug patch BN layers removed dpx = (self.p1 - self.p2) * x return dpx * torch.sigmoid(beta * dpx) + self.p2 * x ================================================ FILE: RStask/ObjectDetection/utils/augmentations.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Image augmentation functions """ import math import random import cv2 import numpy as np from RStask.ObjectDetection.utils.general import LOGGER, check_version, colorstr, resample_segments, segment2box from RStask.ObjectDetection.utils.metrics import bbox_ioa class Albumentations: # YOLOv5 Albumentations class (optional, only used if package is installed) def __init__(self): self.transform = None try: import albumentations as A check_version(A.__version__, '1.0.3', hard=True) # version requirement T = [ A.Blur(p=0.01), A.MedianBlur(p=0.01), A.ToGray(p=0.01), A.CLAHE(p=0.01), A.RandomBrightnessContrast(p=0.0), A.RandomGamma(p=0.0), A.ImageCompression(quality_lower=75, p=0.0)] # transforms self.transform = A.Compose(T, bbox_params=A.BboxParams(format='yolo', label_fields=['class_labels'])) LOGGER.info(colorstr('albumentations: ') + ', '.join(f'{x}' for x in self.transform.transforms if x.p)) except ImportError: # package not installed, skip pass except Exception as e: LOGGER.info(colorstr('albumentations: ') + f'{e}') def __call__(self, im, labels, p=1.0): if self.transform and random.random() < p: new = self.transform(image=im, bboxes=labels[:, 1:], class_labels=labels[:, 0]) # transformed im, labels = new['image'], np.array([[c, *b] for c, b in zip(new['class_labels'], new['bboxes'])]) return im, labels def augment_hsv(im, hgain=0.5, sgain=0.5, vgain=0.5): # HSV color-space augmentation if hgain or sgain or vgain: r = np.random.uniform(-1, 1, 3) * [hgain, sgain, vgain] + 1 # random gains hue, sat, val = cv2.split(cv2.cvtColor(im, cv2.COLOR_BGR2HSV)) dtype = im.dtype # uint8 x = np.arange(0, 256, dtype=r.dtype) lut_hue = ((x * r[0]) % 180).astype(dtype) lut_sat = np.clip(x * r[1], 0, 255).astype(dtype) lut_val = np.clip(x * r[2], 0, 255).astype(dtype) im_hsv = cv2.merge((cv2.LUT(hue, lut_hue), cv2.LUT(sat, lut_sat), cv2.LUT(val, lut_val))) cv2.cvtColor(im_hsv, cv2.COLOR_HSV2BGR, dst=im) # no return needed def hist_equalize(im, clahe=True, bgr=False): # Equalize histogram on BGR image 'im' with im.shape(n,m,3) and range 0-255 yuv = cv2.cvtColor(im, cv2.COLOR_BGR2YUV if bgr else cv2.COLOR_RGB2YUV) if clahe: c = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8)) yuv[:, :, 0] = c.apply(yuv[:, :, 0]) else: yuv[:, :, 0] = cv2.equalizeHist(yuv[:, :, 0]) # equalize Y channel histogram return cv2.cvtColor(yuv, cv2.COLOR_YUV2BGR if bgr else cv2.COLOR_YUV2RGB) # convert YUV image to RGB def replicate(im, labels): # Replicate labels h, w = im.shape[:2] boxes = labels[:, 1:].astype(int) x1, y1, x2, y2 = boxes.T s = ((x2 - x1) + (y2 - y1)) / 2 # side length (pixels) for i in s.argsort()[:round(s.size * 0.5)]: # smallest indices x1b, y1b, x2b, y2b = boxes[i] bh, bw = y2b - y1b, x2b - x1b yc, xc = int(random.uniform(0, h - bh)), int(random.uniform(0, w - bw)) # offset x, y x1a, y1a, x2a, y2a = [xc, yc, xc + bw, yc + bh] im[y1a:y2a, x1a:x2a] = im[y1b:y2b, x1b:x2b] # im4[ymin:ymax, xmin:xmax] labels = np.append(labels, [[labels[i, 0], x1a, y1a, x2a, y2a]], axis=0) return im, labels def letterbox(im, new_shape=(640, 640), color=(114, 114, 114), auto=True, scaleFill=False, scaleup=True, stride=32): # Resize and pad image while meeting stride-multiple constraints shape = im.shape[:2] # current shape [height, width] if isinstance(new_shape, int): new_shape = (new_shape, new_shape) # Scale ratio (new / old) r = min(new_shape[0] / shape[0], new_shape[1] / shape[1]) if not scaleup: # only scale down, do not scale up (for better val mAP) r = min(r, 1.0) # Compute padding ratio = r, r # width, height ratios new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r)) dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1] # wh padding if auto: # minimum rectangle dw, dh = np.mod(dw, stride), np.mod(dh, stride) # wh padding elif scaleFill: # stretch dw, dh = 0.0, 0.0 new_unpad = (new_shape[1], new_shape[0]) ratio = new_shape[1] / shape[1], new_shape[0] / shape[0] # width, height ratios dw /= 2 # divide padding into 2 sides dh /= 2 if shape[::-1] != new_unpad: # resize im = cv2.resize(im, new_unpad, interpolation=cv2.INTER_LINEAR) top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1)) left, right = int(round(dw - 0.1)), int(round(dw + 0.1)) im = cv2.copyMakeBorder(im, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color) # add border return im, ratio, (dw, dh) def random_perspective(im, targets=(), segments=(), degrees=10, translate=.1, scale=.1, shear=10, perspective=0.0, border=(0, 0)): # torchvision.transforms.RandomAffine(degrees=(-10, 10), translate=(0.1, 0.1), scale=(0.9, 1.1), shear=(-10, 10)) # targets = [cls, xyxy] height = im.shape[0] + border[0] * 2 # shape(h,w,c) width = im.shape[1] + border[1] * 2 # Center C = np.eye(3) C[0, 2] = -im.shape[1] / 2 # x translation (pixels) C[1, 2] = -im.shape[0] / 2 # y translation (pixels) # Perspective P = np.eye(3) P[2, 0] = random.uniform(-perspective, perspective) # x perspective (about y) P[2, 1] = random.uniform(-perspective, perspective) # y perspective (about x) # Rotation and Scale R = np.eye(3) a = random.uniform(-degrees, degrees) # a += random.choice([-180, -90, 0, 90]) # add 90deg rotations to small rotations s = random.uniform(1 - scale, 1 + scale) # s = 2 ** random.uniform(-scale, scale) R[:2] = cv2.getRotationMatrix2D(angle=a, center=(0, 0), scale=s) # Shear S = np.eye(3) S[0, 1] = math.tan(random.uniform(-shear, shear) * math.pi / 180) # x shear (deg) S[1, 0] = math.tan(random.uniform(-shear, shear) * math.pi / 180) # y shear (deg) # Translation T = np.eye(3) T[0, 2] = random.uniform(0.5 - translate, 0.5 + translate) * width # x translation (pixels) T[1, 2] = random.uniform(0.5 - translate, 0.5 + translate) * height # y translation (pixels) # Combined rotation matrix M = T @ S @ R @ P @ C # order of operations (right to left) is IMPORTANT if (border[0] != 0) or (border[1] != 0) or (M != np.eye(3)).any(): # image changed if perspective: im = cv2.warpPerspective(im, M, dsize=(width, height), borderValue=(114, 114, 114)) else: # affine im = cv2.warpAffine(im, M[:2], dsize=(width, height), borderValue=(114, 114, 114)) # Visualize # import matplotlib.pyplot as plt # ax = plt.subplots(1, 2, figsize=(12, 6))[1].ravel() # ax[0].imshow(im[:, :, ::-1]) # base # ax[1].imshow(im2[:, :, ::-1]) # warped # Transform label coordinates n = len(targets) if n: use_segments = any(x.any() for x in segments) new = np.zeros((n, 4)) if use_segments: # warp segments segments = resample_segments(segments) # upsample for i, segment in enumerate(segments): xy = np.ones((len(segment), 3)) xy[:, :2] = segment xy = xy @ M.T # transform xy = xy[:, :2] / xy[:, 2:3] if perspective else xy[:, :2] # perspective rescale or affine # clip new[i] = segment2box(xy, width, height) else: # warp boxes xy = np.ones((n * 4, 3)) xy[:, :2] = targets[:, [1, 2, 3, 4, 1, 4, 3, 2]].reshape(n * 4, 2) # x1y1, x2y2, x1y2, x2y1 xy = xy @ M.T # transform xy = (xy[:, :2] / xy[:, 2:3] if perspective else xy[:, :2]).reshape(n, 8) # perspective rescale or affine # create new boxes x = xy[:, [0, 2, 4, 6]] y = xy[:, [1, 3, 5, 7]] new = np.concatenate((x.min(1), y.min(1), x.max(1), y.max(1))).reshape(4, n).T # clip new[:, [0, 2]] = new[:, [0, 2]].clip(0, width) new[:, [1, 3]] = new[:, [1, 3]].clip(0, height) # filter candidates i = box_candidates(box1=targets[:, 1:5].T * s, box2=new.T, area_thr=0.01 if use_segments else 0.10) targets = targets[i] targets[:, 1:5] = new[i] return im, targets def copy_paste(im, labels, segments, p=0.5): # Implement Copy-Paste augmentation https://arxiv.org/abs/2012.07177, labels as nx5 np.array(cls, xyxy) n = len(segments) if p and n: h, w, c = im.shape # height, width, channels im_new = np.zeros(im.shape, np.uint8) for j in random.sample(range(n), k=round(p * n)): l, s = labels[j], segments[j] box = w - l[3], l[2], w - l[1], l[4] ioa = bbox_ioa(box, labels[:, 1:5]) # intersection over area if (ioa < 0.30).all(): # allow 30% obscuration of existing labels labels = np.concatenate((labels, [[l[0], *box]]), 0) segments.append(np.concatenate((w - s[:, 0:1], s[:, 1:2]), 1)) cv2.drawContours(im_new, [segments[j].astype(np.int32)], -1, (255, 255, 255), cv2.FILLED) result = cv2.bitwise_and(src1=im, src2=im_new) result = cv2.flip(result, 1) # augment segments (flip left-right) i = result > 0 # pixels to replace # i[:, :] = result.max(2).reshape(h, w, 1) # act over ch im[i] = result[i] # cv2.imwrite('debug.jpg', im) # debug return im, labels, segments def cutout(im, labels, p=0.5): # Applies image cutout augmentation https://arxiv.org/abs/1708.04552 if random.random() < p: h, w = im.shape[:2] scales = [0.5] * 1 + [0.25] * 2 + [0.125] * 4 + [0.0625] * 8 + [0.03125] * 16 # image size fraction for s in scales: mask_h = random.randint(1, int(h * s)) # create random masks mask_w = random.randint(1, int(w * s)) # box xmin = max(0, random.randint(0, w) - mask_w // 2) ymin = max(0, random.randint(0, h) - mask_h // 2) xmax = min(w, xmin + mask_w) ymax = min(h, ymin + mask_h) # apply random color mask im[ymin:ymax, xmin:xmax] = [random.randint(64, 191) for _ in range(3)] # return unobscured labels if len(labels) and s > 0.03: box = np.array([xmin, ymin, xmax, ymax], dtype=np.float32) ioa = bbox_ioa(box, labels[:, 1:5]) # intersection over area labels = labels[ioa < 0.60] # remove >60% obscured labels return labels def mixup(im, labels, im2, labels2): # Applies MixUp augmentation https://arxiv.org/pdf/1710.09412.pdf r = np.random.beta(32.0, 32.0) # mixup ratio, alpha=beta=32.0 im = (im * r + im2 * (1 - r)).astype(np.uint8) labels = np.concatenate((labels, labels2), 0) return im, labels def box_candidates(box1, box2, wh_thr=2, ar_thr=100, area_thr=0.1, eps=1e-16): # box1(4,n), box2(4,n) # Compute candidate boxes: box1 before augment, box2 after augment, wh_thr (pixels), aspect_ratio_thr, area_ratio w1, h1 = box1[2] - box1[0], box1[3] - box1[1] w2, h2 = box2[2] - box2[0], box2[3] - box2[1] ar = np.maximum(w2 / (h2 + eps), h2 / (w2 + eps)) # aspect ratio return (w2 > wh_thr) & (h2 > wh_thr) & (w2 * h2 / (w1 * h1 + eps) > area_thr) & (ar < ar_thr) # candidates ================================================ FILE: RStask/ObjectDetection/utils/autoanchor.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ AutoAnchor utils """ import random import numpy as np import torch import yaml from tqdm.auto import tqdm from utils.general import LOGGER, colorstr, emojis PREFIX = colorstr('AutoAnchor: ') def check_anchor_order(m): # Check anchor order against stride order for YOLOv5 Detect() module m, and correct if necessary a = m.anchors.prod(-1).mean(-1).view(-1) # mean anchor area per output layer da = a[-1] - a[0] # delta a ds = m.stride[-1] - m.stride[0] # delta s if da and (da.sign() != ds.sign()): # same order LOGGER.info(f'{PREFIX}Reversing anchor order') m.anchors[:] = m.anchors.flip(0) def check_anchors(dataset, model, thr=4.0, imgsz=640): # Check anchor fit to data, recompute if necessary m = model.module.model[-1] if hasattr(model, 'module') else model.model[-1] # Detect() shapes = imgsz * dataset.shapes / dataset.shapes.max(1, keepdims=True) scale = np.random.uniform(0.9, 1.1, size=(shapes.shape[0], 1)) # augment scale wh = torch.tensor(np.concatenate([l[:, 3:5] * s for s, l in zip(shapes * scale, dataset.labels)])).float() # wh def metric(k): # compute metric r = wh[:, None] / k[None] x = torch.min(r, 1 / r).min(2)[0] # ratio metric best = x.max(1)[0] # best_x aat = (x > 1 / thr).float().sum(1).mean() # anchors above threshold bpr = (best > 1 / thr).float().mean() # best possible recall return bpr, aat stride = m.stride.to(m.anchors.device).view(-1, 1, 1) # model strides anchors = m.anchors.clone() * stride # current anchors bpr, aat = metric(anchors.cpu().view(-1, 2)) s = f'\n{PREFIX}{aat:.2f} anchors/target, {bpr:.3f} Best Possible Recall (BPR). ' if bpr > 0.98: # threshold to recompute LOGGER.info(emojis(f'{s}Current anchors are a good fit to dataset ✅')) else: LOGGER.info(emojis(f'{s}Anchors are a poor fit to dataset ⚠️, attempting to improve...')) na = m.anchors.numel() // 2 # number of anchors try: anchors = kmean_anchors(dataset, n=na, img_size=imgsz, thr=thr, gen=1000, verbose=False) except Exception as e: LOGGER.info(f'{PREFIX}ERROR: {e}') new_bpr = metric(anchors)[0] if new_bpr > bpr: # replace anchors anchors = torch.tensor(anchors, device=m.anchors.device).type_as(m.anchors) m.anchors[:] = anchors.clone().view_as(m.anchors) check_anchor_order(m) # must be in pixel-space (not grid-space) m.anchors /= stride s = f'{PREFIX}Done ✅ (optional: update model *.yaml to use these anchors in the future)' else: s = f'{PREFIX}Done ⚠️ (original anchors better than new anchors, proceeding with original anchors)' LOGGER.info(emojis(s)) def kmean_anchors(dataset='./data/coco128.yaml', n=9, img_size=640, thr=4.0, gen=1000, verbose=True): """ Creates kmeans-evolved anchors from training dataset Arguments: dataset: path to data.yaml, or a loaded dataset n: number of anchors img_size: image size used for training thr: anchor-label wh ratio threshold hyperparameter hyp['anchor_t'] used for training, default=4.0 gen: generations to evolve anchors using genetic algorithm verbose: print all results Return: k: kmeans evolved anchors Usage: from utils.autoanchor import *; _ = kmean_anchors() """ from scipy.cluster.vq import kmeans npr = np.random thr = 1 / thr def metric(k, wh): # compute metrics r = wh[:, None] / k[None] x = torch.min(r, 1 / r).min(2)[0] # ratio metric # x = wh_iou(wh, torch.tensor(k)) # iou metric return x, x.max(1)[0] # x, best_x def anchor_fitness(k): # mutation fitness _, best = metric(torch.tensor(k, dtype=torch.float32), wh) return (best * (best > thr).float()).mean() # fitness def print_results(k, verbose=True): k = k[np.argsort(k.prod(1))] # sort small to large x, best = metric(k, wh0) bpr, aat = (best > thr).float().mean(), (x > thr).float().mean() * n # best possible recall, anch > thr s = f'{PREFIX}thr={thr:.2f}: {bpr:.4f} best possible recall, {aat:.2f} anchors past thr\n' \ f'{PREFIX}n={n}, img_size={img_size}, metric_all={x.mean():.3f}/{best.mean():.3f}-mean/best, ' \ f'past_thr={x[x > thr].mean():.3f}-mean: ' for i, x in enumerate(k): s += '%i,%i, ' % (round(x[0]), round(x[1])) if verbose: LOGGER.info(s[:-2]) return k if isinstance(dataset, str): # *.yaml file with open(dataset, errors='ignore') as f: data_dict = yaml.safe_load(f) # model dict from utils.datasets import LoadImagesAndLabels dataset = LoadImagesAndLabels(data_dict['train'], augment=True, rect=True) # Get label wh shapes = img_size * dataset.shapes / dataset.shapes.max(1, keepdims=True) wh0 = np.concatenate([l[:, 3:5] * s for s, l in zip(shapes, dataset.labels)]) # wh # Filter i = (wh0 < 3.0).any(1).sum() if i: LOGGER.info(f'{PREFIX}WARNING: Extremely small objects found: {i} of {len(wh0)} labels are < 3 pixels in size') wh = wh0[(wh0 >= 2.0).any(1)] # filter > 2 pixels # wh = wh * (npr.rand(wh.shape[0], 1) * 0.9 + 0.1) # multiply by random scale 0-1 # Kmeans init try: LOGGER.info(f'{PREFIX}Running kmeans for {n} anchors on {len(wh)} points...') assert n <= len(wh) # apply overdetermined constraint s = wh.std(0) # sigmas for whitening k = kmeans(wh / s, n, iter=30)[0] * s # points assert n == len(k) # kmeans may return fewer points than requested if wh is insufficient or too similar except Exception: LOGGER.warning(f'{PREFIX}WARNING: switching strategies from kmeans to random init') k = np.sort(npr.rand(n * 2)).reshape(n, 2) * img_size # random init wh, wh0 = (torch.tensor(x, dtype=torch.float32) for x in (wh, wh0)) k = print_results(k, verbose=False) # Plot # k, d = [None] * 20, [None] * 20 # for i in tqdm(range(1, 21)): # k[i-1], d[i-1] = kmeans(wh / s, i) # points, mean distance # fig, ax = plt.subplots(1, 2, figsize=(14, 7), tight_layout=True) # ax = ax.ravel() # ax[0].plot(np.arange(1, 21), np.array(d) ** 2, marker='.') # fig, ax = plt.subplots(1, 2, figsize=(14, 7)) # plot wh # ax[0].hist(wh[wh[:, 0]<100, 0],400) # ax[1].hist(wh[wh[:, 1]<100, 1],400) # fig.savefig('wh.png', dpi=200) # Evolve f, sh, mp, s = anchor_fitness(k), k.shape, 0.9, 0.1 # fitness, generations, mutation prob, sigma pbar = tqdm(range(gen), bar_format='{l_bar}{bar:10}{r_bar}{bar:-10b}') # progress bar for _ in pbar: v = np.ones(sh) while (v == 1).all(): # mutate until a change occurs (prevent duplicates) v = ((npr.random(sh) < mp) * random.random() * npr.randn(*sh) * s + 1).clip(0.3, 3.0) kg = (k.copy() * v).clip(min=2.0) fg = anchor_fitness(kg) if fg > f: f, k = fg, kg.copy() pbar.desc = f'{PREFIX}Evolving anchors with Genetic Algorithm: fitness = {f:.4f}' if verbose: print_results(k, verbose) return print_results(k) ================================================ FILE: RStask/ObjectDetection/utils/autobatch.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Auto-batch utils """ from copy import deepcopy import numpy as np import torch from torch.cuda import amp from utils.general import LOGGER, colorstr from utils.torch_utils import profile def check_train_batch_size(model, imgsz=640): # Check YOLOv5 training batch size with amp.autocast(): return autobatch(deepcopy(model).train(), imgsz) # compute optimal batch size def autobatch(model, imgsz=640, fraction=0.9, batch_size=16): # Automatically estimate best batch size to use `fraction` of available CUDA memory # Usage: # import torch # from utils.autobatch import autobatch # model = torch.hub.load('ultralytics/yolov5', 'yolov5s', autoshape=False) # print(autobatch(model)) prefix = colorstr('AutoBatch: ') LOGGER.info(f'{prefix}Computing optimal batch size for --imgsz {imgsz}') device = next(model.parameters()).device # get model device if device.type == 'cpu': LOGGER.info(f'{prefix}CUDA not detected, using default CPU batch-size {batch_size}') return batch_size gb = 1 << 30 # bytes to GiB (1024 ** 3) d = str(device).upper() # 'CUDA:0' properties = torch.cuda.get_device_properties(device) # device properties t = properties.total_memory / gb # (GiB) r = torch.cuda.memory_reserved(device) / gb # (GiB) a = torch.cuda.memory_allocated(device) / gb # (GiB) f = t - (r + a) # free inside reserved LOGGER.info(f'{prefix}{d} ({properties.name}) {t:.2f}G total, {r:.2f}G reserved, {a:.2f}G allocated, {f:.2f}G free') batch_sizes = [1, 2, 4, 8, 16] try: img = [torch.zeros(b, 3, imgsz, imgsz) for b in batch_sizes] y = profile(img, model, n=3, device=device) except Exception as e: LOGGER.warning(f'{prefix}{e}') y = [x[2] for x in y if x] # memory [2] batch_sizes = batch_sizes[:len(y)] p = np.polyfit(batch_sizes, y, deg=1) # first degree polynomial fit b = int((f * fraction - p[1]) / p[0]) # y intercept (optimal batch size) LOGGER.info(f'{prefix}Using batch-size {b} for {d} {t * fraction:.2f}G/{t:.2f}G ({fraction * 100:.0f}%)') return b ================================================ FILE: RStask/ObjectDetection/utils/aws/__init__.py ================================================ ================================================ FILE: RStask/ObjectDetection/utils/aws/mime.sh ================================================ # AWS EC2 instance startup 'MIME' script https://aws.amazon.com/premiumsupport/knowledge-center/execute-user-data-ec2/ # This script will run on every instance restart, not only on first start # --- DO NOT COPY ABOVE COMMENTS WHEN PASTING INTO USERDATA --- Content-Type: multipart/mixed; boundary="//" MIME-Version: 1.0 --// Content-Type: text/cloud-config; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="cloud-config.txt" #cloud-config cloud_final_modules: - [scripts-user, always] --// Content-Type: text/x-shellscript; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="userdata.txt" #!/bin/bash # --- paste contents of userdata.sh here --- --// ================================================ FILE: RStask/ObjectDetection/utils/aws/resume.py ================================================ # Resume all interrupted trainings in yolov5/ dir including DDP trainings # Usage: $ python utils/aws/resume.py import os import sys from pathlib import Path import torch import yaml FILE = Path(__file__).resolve() ROOT = FILE.parents[2] # YOLOv5 root directory if str(ROOT) not in sys.path: sys.path.append(str(ROOT)) # add ROOT to PATH port = 0 # --master_port path = Path('').resolve() for last in path.rglob('*/**/last.pt'): ckpt = torch.load(last) if ckpt['optimizer'] is None: continue # Load opt.yaml with open(last.parent.parent / 'opt.yaml', errors='ignore') as f: opt = yaml.safe_load(f) # Get device count d = opt['device'].split(',') # devices nd = len(d) # number of devices ddp = nd > 1 or (nd == 0 and torch.cuda.device_count() > 1) # distributed data parallel if ddp: # multi-GPU port += 1 cmd = f'python -m torch.distributed.run --nproc_per_node {nd} --master_port {port} train.py --resume {last}' else: # single-GPU cmd = f'python train.py --resume {last}' cmd += ' > /dev/null 2>&1 &' # redirect output to dev/null and run in daemon thread print(cmd) os.system(cmd) ================================================ FILE: RStask/ObjectDetection/utils/aws/userdata.sh ================================================ #!/bin/bash # AWS EC2 instance startup script https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/user-data.html # This script will run only once on first instance start (for a re-start script see mime.sh) # /home/ubuntu (ubuntu) or /home/ec2-user (amazon-linux) is working dir # Use >300 GB SSD cd home/ubuntu if [ ! -d yolov5 ]; then echo "Running first-time script." # install dependencies, download COCO, pull Docker git clone https://github.com/ultralytics/yolov5 -b master && sudo chmod -R 777 yolov5 cd yolov5 bash data/scripts/get_coco.sh && echo "COCO done." & sudo docker pull ultralytics/yolov5:latest && echo "Docker done." & python -m pip install --upgrade pip && pip install -r requirements.txt && python detect.py && echo "Requirements done." & wait && echo "All tasks done." # finish background tasks else echo "Running re-start script." # resume interrupted runs i=0 list=$(sudo docker ps -qa) # container list i.e. $'one\ntwo\nthree\nfour' while IFS= read -r id; do ((i++)) echo "restarting container $i: $id" sudo docker start $id # sudo docker exec -it $id python train.py --resume # single-GPU sudo docker exec -d $id python utils/aws/resume.py # multi-scenario done <<<"$list" fi ================================================ FILE: RStask/ObjectDetection/utils/benchmarks.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Run YOLOv5 benchmarks on all supported export formats Format | `export.py --include` | Model --- | --- | --- PyTorch | - | yolov5s.pt TorchScript | `torchscript` | yolov5s.torchscript ONNX | `onnx` | yolov5s.onnx OpenVINO | `openvino` | yolov5s_openvino_model/ TensorRT | `engine` | yolov5s.engine CoreML | `coreml` | yolov5s.mlmodel TensorFlow SavedModel | `saved_model` | yolov5s_saved_model/ TensorFlow GraphDef | `pb` | yolov5s.pb TensorFlow Lite | `tflite` | yolov5s.tflite TensorFlow Edge TPU | `edgetpu` | yolov5s_edgetpu.tflite TensorFlow.js | `tfjs` | yolov5s_web_model/ Requirements: $ pip install -r requirements.txt coremltools onnx onnx-simplifier onnxruntime openvino-dev tensorflow-cpu # CPU $ pip install -r requirements.txt coremltools onnx onnx-simplifier onnxruntime-gpu openvino-dev tensorflow # GPU $ pip install -U nvidia-tensorrt --index-url https://pypi.ngc.nvidia.com # TensorRT Usage: $ python utils/benchmarks.py --weights yolov5s.pt --img 640 """ import argparse import sys import time from pathlib import Path import pandas as pd FILE = Path(__file__).resolve() ROOT = FILE.parents[1] # YOLOv5 root directory if str(ROOT) not in sys.path: sys.path.append(str(ROOT)) # add ROOT to PATH # ROOT = ROOT.relative_to(Path.cwd()) # relative import export import val from utils import notebook_init from utils.general import LOGGER, print_args from utils.torch_utils import select_device def run( weights=ROOT / 'yolov5s.pt', # weights path imgsz=640, # inference size (pixels) batch_size=1, # batch size data=ROOT / 'data/coco128.yaml', # dataset.yaml path device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu half=False, # use FP16 half-precision inference test=False, # test exports only pt_only=False, # test PyTorch only ): y, t = [], time.time() formats = export.export_formats() device = select_device(device) for i, (name, f, suffix, gpu) in formats.iterrows(): # index, (name, file, suffix, gpu-capable) try: assert i != 9, 'Edge TPU not supported' assert i != 10, 'TF.js not supported' if device.type != 'cpu': assert gpu, f'{name} inference not supported on GPU' # Export if f == '-': w = weights # PyTorch format else: w = export.run(weights=weights, imgsz=[imgsz], include=[f], device=device, half=half)[-1] # all others assert suffix in str(w), 'export failed' # Validate result = val.run(data, w, batch_size, imgsz, plots=False, device=device, task='benchmark', half=half) metrics = result[0] # metrics (mp, mr, map50, map, *losses(box, obj, cls)) speeds = result[2] # times (preprocess, inference, postprocess) y.append([name, round(metrics[3], 4), round(speeds[1], 2)]) # mAP, t_inference except Exception as e: LOGGER.warning(f'WARNING: Benchmark failure for {name}: {e}') y.append([name, None, None]) # mAP, t_inference if pt_only and i == 0: break # break after PyTorch # Print results LOGGER.info('\n') parse_opt() notebook_init() # print system info py = pd.DataFrame(y, columns=['Format', 'mAP@0.5:0.95', 'Inference time (ms)'] if map else ['Format', 'Export', '']) LOGGER.info(f'\nBenchmarks complete ({time.time() - t:.2f}s)') LOGGER.info(str(py if map else py.iloc[:, :2])) return py def test( weights=ROOT / 'yolov5s.pt', # weights path imgsz=640, # inference size (pixels) batch_size=1, # batch size data=ROOT / 'data/coco128.yaml', # dataset.yaml path device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu half=False, # use FP16 half-precision inference test=False, # test exports only pt_only=False, # test PyTorch only ): y, t = [], time.time() formats = export.export_formats() device = select_device(device) for i, (name, f, suffix, gpu) in formats.iterrows(): # index, (name, file, suffix, gpu-capable) try: w = weights if f == '-' else \ export.run(weights=weights, imgsz=[imgsz], include=[f], device=device, half=half)[-1] # weights assert suffix in str(w), 'export failed' y.append([name, True]) except Exception: y.append([name, False]) # mAP, t_inference # Print results LOGGER.info('\n') parse_opt() notebook_init() # print system info py = pd.DataFrame(y, columns=['Format', 'Export']) LOGGER.info(f'\nExports complete ({time.time() - t:.2f}s)') LOGGER.info(str(py)) return py def parse_opt(): parser = argparse.ArgumentParser() parser.add_argument('--weights', type=str, default=ROOT / 'result_300epoch/best.pt', help='weights path') parser.add_argument('--imgsz', '--img', '--img-size', type=int, default=640, help='inference size (pixels)') parser.add_argument('--batch-size', type=int, default=1, help='batch size') parser.add_argument('--data', type=str, default=ROOT / 'data_data/dota_name.yaml', help='dataset.yaml path') parser.add_argument('--device', default='1', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') parser.add_argument('--half', action='store_true', help='use FP16 half-precision inference') parser.add_argument('--test', action='store_true', help='test exports only') parser.add_argument('--pt-only', action='store_true', help='test PyTorch only') opt = parser.parse_args() print_args(vars(opt)) return opt def main(opt): test(**vars(opt)) if opt.test else run(**vars(opt)) if __name__ == "__main__": opt = parse_opt() main(opt) ================================================ FILE: RStask/ObjectDetection/utils/callbacks.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Callback utils """ class Callbacks: """" Handles all registered callbacks for YOLOv5 Hooks """ def __init__(self): # Define the available callbacks self._callbacks = { 'on_pretrain_routine_start': [], 'on_pretrain_routine_end': [], 'on_train_start': [], 'on_train_epoch_start': [], 'on_train_batch_start': [], 'optimizer_step': [], 'on_before_zero_grad': [], 'on_train_batch_end': [], 'on_train_epoch_end': [], 'on_val_start': [], 'on_val_batch_start': [], 'on_val_image_end': [], 'on_val_batch_end': [], 'on_val_end': [], 'on_fit_epoch_end': [], # fit = train + val 'on_model_save': [], 'on_train_end': [], 'on_params_update': [], 'teardown': [],} self.stop_training = False # set True to interrupt training def register_action(self, hook, name='', callback=None): """ Register a new action to a callback hook Args: hook: The callback hook name to register the action to name: The name of the action for later reference callback: The callback to fire """ assert hook in self._callbacks, f"hook '{hook}' not found in callbacks {self._callbacks}" assert callable(callback), f"callback '{callback}' is not callable" self._callbacks[hook].append({'name': name, 'callback': callback}) def get_registered_actions(self, hook=None): """" Returns all the registered actions by callback hook Args: hook: The name of the hook to check, defaults to all """ return self._callbacks[hook] if hook else self._callbacks def run(self, hook, *args, **kwargs): """ Loop through the registered actions and fire all callbacks Args: hook: The name of the hook to check, defaults to all args: Arguments to receive from YOLOv5 kwargs: Keyword Arguments to receive from YOLOv5 """ assert hook in self._callbacks, f"hook '{hook}' not found in callbacks {self._callbacks}" for logger in self._callbacks[hook]: logger['callback'](*args, **kwargs) ================================================ FILE: RStask/ObjectDetection/utils/datasets.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Dataloaders and dataset utils """ import glob import hashlib import json import math import os import random import shutil import time from itertools import repeat from multiprocessing.pool import Pool, ThreadPool from pathlib import Path from threading import Thread from urllib.parse import urlparse from zipfile import ZipFile import numpy as np import torch import torch.nn.functional as F import yaml from PIL import ExifTags, Image, ImageOps from torch.utils.data import DataLoader, Dataset, dataloader, distributed from tqdm.auto import tqdm from RStask.ObjectDetection.utils.augmentations import Albumentations, augment_hsv, copy_paste, letterbox, mixup, random_perspective from RStask.ObjectDetection.utils.general import (DATASETS_DIR, LOGGER, NUM_THREADS, check_dataset, check_requirements, check_yaml, clean_str, cv2, segments2boxes, xyn2xy, xywh2xyxy, xywhn2xyxy, xyxy2xywhn) from RStask.ObjectDetection.utils.torch_utils import torch_distributed_zero_first # Parameters HELP_URL = 'https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data' IMG_FORMATS = 'bmp', 'dng', 'jpeg', 'jpg', 'mpo', 'png', 'tif', 'tiff', 'webp' # include image suffixes VID_FORMATS = 'asf', 'avi', 'gif', 'm4v', 'mkv', 'mov', 'mp4', 'mpeg', 'mpg', 'ts', 'wmv' # include video suffixes BAR_FORMAT = '{l_bar}{bar:10}{r_bar}{bar:-10b}' # tqdm bar format LOCAL_RANK = int(os.getenv('LOCAL_RANK', -1)) # https://pytorch.org/docs/stable/elastic/run.html # Get orientation exif tag for orientation in ExifTags.TAGS.keys(): if ExifTags.TAGS[orientation] == 'Orientation': break def get_hash(paths): # Returns a single hash value of a list of paths (files or dirs) size = sum(os.path.getsize(p) for p in paths if os.path.exists(p)) # sizes h = hashlib.md5(str(size).encode()) # hash sizes h.update(''.join(paths).encode()) # hash paths return h.hexdigest() # return hash def exif_size(img): # Returns exif-corrected PIL size s = img.size # (width, height) try: rotation = dict(img._getexif().items())[orientation] if rotation == 6: # rotation 270 s = (s[1], s[0]) elif rotation == 8: # rotation 90 s = (s[1], s[0]) except Exception: pass return s def exif_transpose(image): """ Transpose a PIL image accordingly if it has an EXIF Orientation tag. Inplace version of https://github.com/python-pillow/Pillow/blob/master/src/PIL/ImageOps.py exif_transpose() :param image: The image to transpose. :return: An image. """ exif = image.getexif() orientation = exif.get(0x0112, 1) # default 1 if orientation > 1: method = { 2: Image.FLIP_LEFT_RIGHT, 3: Image.ROTATE_180, 4: Image.FLIP_TOP_BOTTOM, 5: Image.TRANSPOSE, 6: Image.ROTATE_270, 7: Image.TRANSVERSE, 8: Image.ROTATE_90,}.get(orientation) if method is not None: image = image.transpose(method) del exif[0x0112] image.info["exif"] = exif.tobytes() return image def create_dataloader(path, imgsz, batch_size, stride, single_cls=False, hyp=None, augment=False, cache=False, pad=0.0, rect=False, rank=-1, workers=8, image_weights=False, quad=False, prefix='', shuffle=False): if rect and shuffle: LOGGER.warning('WARNING: --rect is incompatible with DataLoader shuffle, setting shuffle=False') shuffle = False with torch_distributed_zero_first(rank): # init dataset *.cache only once if DDP dataset = LoadImagesAndLabels( path, imgsz, batch_size, augment=augment, # augmentation hyp=hyp, # hyperparameters rect=rect, # rectangular batches cache_images=cache, single_cls=single_cls, stride=int(stride), pad=pad, image_weights=image_weights, prefix=prefix) batch_size = min(batch_size, len(dataset)) nd = torch.cuda.device_count() # number of CUDA devices nw = min([os.cpu_count() // max(nd, 1), batch_size if batch_size > 1 else 0, workers]) # number of workers sampler = None if rank == -1 else distributed.DistributedSampler(dataset, shuffle=shuffle) loader = DataLoader if image_weights else InfiniteDataLoader # only DataLoader allows for attribute updates return loader(dataset, batch_size=batch_size, shuffle=shuffle and sampler is None, num_workers=nw, sampler=sampler, pin_memory=True, collate_fn=LoadImagesAndLabels.collate_fn4 if quad else LoadImagesAndLabels.collate_fn), dataset class InfiniteDataLoader(dataloader.DataLoader): """ Dataloader that reuses workers Uses same syntax as vanilla DataLoader """ def __init__(self, *args, **kwargs): super().__init__(*args, **kwargs) object.__setattr__(self, 'batch_sampler', _RepeatSampler(self.batch_sampler)) self.iterator = super().__iter__() def __len__(self): return len(self.batch_sampler.sampler) def __iter__(self): for i in range(len(self)): yield next(self.iterator) class _RepeatSampler: """ Sampler that repeats forever Args: sampler (Sampler) """ def __init__(self, sampler): self.sampler = sampler def __iter__(self): while True: yield from iter(self.sampler) class LoadImages: # YOLOv5 image/video dataloader, i.e. `python detect.py --source image.jpg/vid.mp4` def __init__(self, path, img_size=640, stride=32, auto=True): p = str(Path(path).resolve()) # os-agnostic absolute path if '*' in p: files = sorted(glob.glob(p, recursive=True)) # glob elif os.path.isdir(p): files = sorted(glob.glob(os.path.join(p, '*.*'))) # dir elif os.path.isfile(p): files = [p] # files else: raise Exception(f'ERROR: {p} does not exist') images = [x for x in files if x.split('.')[-1].lower() in IMG_FORMATS] videos = [x for x in files if x.split('.')[-1].lower() in VID_FORMATS] ni, nv = len(images), len(videos) self.img_size = img_size self.stride = stride self.files = images + videos self.nf = ni + nv # number of files self.video_flag = [False] * ni + [True] * nv self.mode = 'image' self.auto = auto if any(videos): self.new_video(videos[0]) # new video else: self.cap = None assert self.nf > 0, f'No images or videos found in {p}. ' \ f'Supported formats are:\nimages: {IMG_FORMATS}\nvideos: {VID_FORMATS}' def __iter__(self): self.count = 0 return self def __next__(self): if self.count == self.nf: raise StopIteration path = self.files[self.count] if self.video_flag[self.count]: # Read video self.mode = 'video' ret_val, img0 = self.cap.read() while not ret_val: self.count += 1 self.cap.release() if self.count == self.nf: # last video raise StopIteration else: path = self.files[self.count] self.new_video(path) ret_val, img0 = self.cap.read() self.frame += 1 s = f'video {self.count + 1}/{self.nf} ({self.frame}/{self.frames}) {path}: ' else: # Read image self.count += 1 img0 = cv2.imread(path) # BGR assert img0 is not None, f'Image Not Found {path}' s = f'image {self.count}/{self.nf} {path}: ' # Padded resize img = letterbox(img0, self.img_size, stride=self.stride, auto=self.auto)[0] # Convert img = img.transpose((2, 0, 1))[::-1] # HWC to CHW, BGR to RGB img = np.ascontiguousarray(img) return path, img, img0, self.cap, s def new_video(self, path): self.frame = 0 self.cap = cv2.VideoCapture(path) self.frames = int(self.cap.get(cv2.CAP_PROP_FRAME_COUNT)) def __len__(self): return self.nf # number of files class LoadWebcam: # for inference # YOLOv5 local webcam dataloader, i.e. `python detect.py --source 0` def __init__(self, pipe='0', img_size=640, stride=32): self.img_size = img_size self.stride = stride self.pipe = eval(pipe) if pipe.isnumeric() else pipe self.cap = cv2.VideoCapture(self.pipe) # video capture object self.cap.set(cv2.CAP_PROP_BUFFERSIZE, 3) # set buffer size def __iter__(self): self.count = -1 return self def __next__(self): self.count += 1 if cv2.waitKey(1) == ord('q'): # q to quit self.cap.release() cv2.destroyAllWindows() raise StopIteration # Read frame ret_val, img0 = self.cap.read() img0 = cv2.flip(img0, 1) # flip left-right # Print assert ret_val, f'Camera Error {self.pipe}' img_path = 'webcam.jpg' s = f'webcam {self.count}: ' # Padded resize img = letterbox(img0, self.img_size, stride=self.stride)[0] # Convert img = img.transpose((2, 0, 1))[::-1] # HWC to CHW, BGR to RGB img = np.ascontiguousarray(img) return img_path, img, img0, None, s def __len__(self): return 0 class LoadStreams: # YOLOv5 streamloader, i.e. `python detect.py --source 'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP streams` def __init__(self, sources='streams.txt', img_size=640, stride=32, auto=True): self.mode = 'stream' self.img_size = img_size self.stride = stride if os.path.isfile(sources): with open(sources) as f: sources = [x.strip() for x in f.read().strip().splitlines() if len(x.strip())] else: sources = [sources] n = len(sources) self.imgs, self.fps, self.frames, self.threads = [None] * n, [0] * n, [0] * n, [None] * n self.sources = [clean_str(x) for x in sources] # clean source names for later self.auto = auto for i, s in enumerate(sources): # index, source # Start thread to read frames from video stream st = f'{i + 1}/{n}: {s}... ' if urlparse(s).hostname in ('www.youtube.com', 'youtube.com', 'youtu.be'): # if source is YouTube video check_requirements(('pafy', 'youtube_dl==2020.12.2')) import pafy s = pafy.new(s).getbest(preftype="mp4").url # YouTube URL s = eval(s) if s.isnumeric() else s # i.e. s = '0' local webcam cap = cv2.VideoCapture(s) assert cap.isOpened(), f'{st}Failed to open {s}' w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)) h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) fps = cap.get(cv2.CAP_PROP_FPS) # warning: may return 0 or nan self.frames[i] = max(int(cap.get(cv2.CAP_PROP_FRAME_COUNT)), 0) or float('inf') # infinite stream fallback self.fps[i] = max((fps if math.isfinite(fps) else 0) % 100, 0) or 30 # 30 FPS fallback _, self.imgs[i] = cap.read() # guarantee first frame self.threads[i] = Thread(target=self.update, args=([i, cap, s]), daemon=True) LOGGER.info(f"{st} Success ({self.frames[i]} frames {w}x{h} at {self.fps[i]:.2f} FPS)") self.threads[i].start() LOGGER.info('') # newline # check for common shapes s = np.stack([letterbox(x, self.img_size, stride=self.stride, auto=self.auto)[0].shape for x in self.imgs]) self.rect = np.unique(s, axis=0).shape[0] == 1 # rect inference if all shapes equal if not self.rect: LOGGER.warning('WARNING: Stream shapes differ. For optimal performance supply similarly-shaped streams.') def update(self, i, cap, stream): # Read stream `i` frames in daemon thread n, f, read = 0, self.frames[i], 1 # frame number, frame array, inference every 'read' frame while cap.isOpened() and n < f: n += 1 # _, self.imgs[index] = cap.read() cap.grab() if n % read == 0: success, im = cap.retrieve() if success: self.imgs[i] = im else: LOGGER.warning('WARNING: Video stream unresponsive, please check your IP camera connection.') self.imgs[i] = np.zeros_like(self.imgs[i]) cap.open(stream) # re-open stream if signal was lost time.sleep(1 / self.fps[i]) # wait time def __iter__(self): self.count = -1 return self def __next__(self): self.count += 1 if not all(x.is_alive() for x in self.threads) or cv2.waitKey(1) == ord('q'): # q to quit cv2.destroyAllWindows() raise StopIteration # Letterbox img0 = self.imgs.copy() img = [letterbox(x, self.img_size, stride=self.stride, auto=self.rect and self.auto)[0] for x in img0] # Stack img = np.stack(img, 0) # Convert img = img[..., ::-1].transpose((0, 3, 1, 2)) # BGR to RGB, BHWC to BCHW img = np.ascontiguousarray(img) return self.sources, img, img0, None, '' def __len__(self): return len(self.sources) # 1E12 frames = 32 streams at 30 FPS for 30 years def img2label_paths(img_paths): # Define label paths as a function of image paths sa, sb = os.sep + 'images' + os.sep, os.sep + 'labels' + os.sep # /images/, /labels/ substrings return [sb.join(x.rsplit(sa, 1)).rsplit('.', 1)[0] + '.txt' for x in img_paths] class LoadImagesAndLabels(Dataset): # YOLOv5 train_loader/val_loader, loads images and labels for training and validation cache_version = 0.6 # dataset labels *.cache version def __init__(self, path, img_size=640, batch_size=16, augment=False, hyp=None, rect=False, image_weights=False, cache_images=False, single_cls=False, stride=32, pad=0.0, prefix=''): self.img_size = img_size self.augment = augment self.hyp = hyp self.image_weights = image_weights self.rect = False if image_weights else rect self.mosaic = self.augment and not self.rect # load 4 images at a time into a mosaic (only during training) self.mosaic_border = [-img_size // 2, -img_size // 2] self.stride = stride self.path = path self.albumentations = Albumentations() if augment else None try: f = [] # image files for p in path if isinstance(path, list) else [path]: p = Path(p) # os-agnostic if p.is_dir(): # dir f += glob.glob(str(p / '**' / '*.*'), recursive=True) # f = list(p.rglob('*.*')) # pathlib elif p.is_file(): # file with open(p) as t: t = t.read().strip().splitlines() parent = str(p.parent) + os.sep f += [x.replace('./', parent) if x.startswith('./') else x for x in t] # local to global path # f += [p.parent / x.lstrip(os.sep) for x in t] # local to global path (pathlib) else: raise Exception(f'{prefix}{p} does not exist') self.im_files = sorted(x.replace('/', os.sep) for x in f if x.split('.')[-1].lower() in IMG_FORMATS) # self.img_files = sorted([x for x in f if x.suffix[1:].lower() in IMG_FORMATS]) # pathlib assert self.im_files, f'{prefix}No images found' except Exception as e: raise Exception(f'{prefix}Error loading data from {path}: {e}\nSee {HELP_URL}') # Check cache self.label_files = img2label_paths(self.im_files) # labels cache_path = (p if p.is_file() else Path(self.label_files[0]).parent).with_suffix('.cache') try: cache, exists = np.load(cache_path, allow_pickle=True).item(), True # load dict assert cache['version'] == self.cache_version # same version assert cache['hash'] == get_hash(self.label_files + self.im_files) # same hash except Exception: cache, exists = self.cache_labels(cache_path, prefix), False # cache # Display cache nf, nm, ne, nc, n = cache.pop('results') # found, missing, empty, corrupt, total if exists and LOCAL_RANK in (-1, 0): d = f"Scanning '{cache_path}' images and labels... {nf} found, {nm} missing, {ne} empty, {nc} corrupt" tqdm(None, desc=prefix + d, total=n, initial=n, bar_format=BAR_FORMAT) # display cache results if cache['msgs']: LOGGER.info('\n'.join(cache['msgs'])) # display warnings assert nf > 0 or not augment, f'{prefix}No labels in {cache_path}. Can not train without labels. See {HELP_URL}' # Read cache [cache.pop(k) for k in ('hash', 'version', 'msgs')] # remove items labels, shapes, self.segments = zip(*cache.values()) self.labels = list(labels) self.shapes = np.array(shapes, dtype=np.float64) self.im_files = list(cache.keys()) # update self.label_files = img2label_paths(cache.keys()) # update n = len(shapes) # number of images bi = np.floor(np.arange(n) / batch_size).astype(np.int32) # batch index nb = bi[-1] + 1 # number of batches self.batch = bi # batch index of image self.n = n self.indices = range(n) # Update labels include_class = [] # filter labels to include only these classes (optional) include_class_array = np.array(include_class).reshape(1, -1) for i, (label, segment) in enumerate(zip(self.labels, self.segments)): if include_class: j = (label[:, 0:1] == include_class_array).any(1) self.labels[i] = label[j] if segment: self.segments[i] = segment[j] if single_cls: # single-class training, merge all classes into 0 self.labels[i][:, 0] = 0 if segment: self.segments[i][:, 0] = 0 # Rectangular Training if self.rect: # Sort by aspect ratio s = self.shapes # wh ar = s[:, 1] / s[:, 0] # aspect ratio irect = ar.argsort() self.im_files = [self.im_files[i] for i in irect] self.label_files = [self.label_files[i] for i in irect] self.labels = [self.labels[i] for i in irect] self.shapes = s[irect] # wh ar = ar[irect] # Set training image shapes shapes = [[1, 1]] * nb for i in range(nb): ari = ar[bi == i] mini, maxi = ari.min(), ari.max() if maxi < 1: shapes[i] = [maxi, 1] elif mini > 1: shapes[i] = [1, 1 / mini] self.batch_shapes = np.ceil(np.array(shapes) * img_size / stride + pad).astype(np.int32) * stride # Cache images into RAM/disk for faster training (WARNING: large datasets may exceed system resources) self.ims = [None] * n self.npy_files = [Path(f).with_suffix('.npy') for f in self.im_files] if cache_images: gb = 0 # Gigabytes of cached images self.im_hw0, self.im_hw = [None] * n, [None] * n fcn = self.cache_images_to_disk if cache_images == 'disk' else self.load_image results = ThreadPool(NUM_THREADS).imap(fcn, range(n)) pbar = tqdm(enumerate(results), total=n, bar_format=BAR_FORMAT, disable=LOCAL_RANK > 0) for i, x in pbar: if cache_images == 'disk': gb += self.npy_files[i].stat().st_size else: # 'ram' self.ims[i], self.im_hw0[i], self.im_hw[i] = x # im, hw_orig, hw_resized = load_image(self, i) gb += self.ims[i].nbytes pbar.desc = f'{prefix}Caching images ({gb / 1E9:.1f}GB {cache_images})' pbar.close() def cache_labels(self, path=Path('./labels.cache'), prefix=''): # Cache dataset labels, check images and read shapes x = {} # dict nm, nf, ne, nc, msgs = 0, 0, 0, 0, [] # number missing, found, empty, corrupt, messages desc = f"{prefix}Scanning '{path.parent / path.stem}' images and labels..." with Pool(NUM_THREADS) as pool: pbar = tqdm(pool.imap(verify_image_label, zip(self.im_files, self.label_files, repeat(prefix))), desc=desc, total=len(self.im_files), bar_format=BAR_FORMAT) for im_file, lb, shape, segments, nm_f, nf_f, ne_f, nc_f, msg in pbar: nm += nm_f nf += nf_f ne += ne_f nc += nc_f if im_file: x[im_file] = [lb, shape, segments] if msg: msgs.append(msg) pbar.desc = f"{desc}{nf} found, {nm} missing, {ne} empty, {nc} corrupt" pbar.close() if msgs: LOGGER.info('\n'.join(msgs)) if nf == 0: LOGGER.warning(f'{prefix}WARNING: No labels found in {path}. See {HELP_URL}') x['hash'] = get_hash(self.label_files + self.im_files) x['results'] = nf, nm, ne, nc, len(self.im_files) x['msgs'] = msgs # warnings x['version'] = self.cache_version # cache version try: np.save(path, x) # save cache for next time path.with_suffix('.cache.npy').rename(path) # remove .npy suffix LOGGER.info(f'{prefix}New cache created: {path}') except Exception as e: LOGGER.warning(f'{prefix}WARNING: Cache directory {path.parent} is not writeable: {e}') # not writeable return x def __len__(self): return len(self.im_files) # def __iter__(self): # self.count = -1 # print('ran dataset iter') # #self.shuffled_vector = np.random.permutation(self.nF) if self.augment else np.arange(self.nF) # return self def __getitem__(self, index): index = self.indices[index] # linear, shuffled, or image_weights hyp = self.hyp mosaic = self.mosaic and random.random() < hyp['mosaic'] img, (h0, w0), (h, w) = self.load_image(index) # Letterbox shape = self.batch_shapes[self.batch[index]] if self.rect else self.img_size # final letterboxed shape img, ratio, pad = letterbox(img, shape, auto=False, scaleup=self.augment) shapes = (h0, w0), ((h / h0, w / w0), pad) # for COCO mAP rescaling labels = self.labels[index].copy() if labels.size: # normalized xywh to pixel xyxy format labels[:, 1:] = xywhn2xyxy(labels[:, 1:], ratio[0] * w, ratio[1] * h, padw=pad[0], padh=pad[1]) nl = len(labels) # number of labels if nl: labels[:, 1:5] = xyxy2xywhn(labels[:, 1:5], w=img.shape[1], h=img.shape[0], clip=True, eps=1E-3) if self.augment: # Albumentations img, labels = self.albumentations(img, labels) nl = len(labels) # update after albumentations # HSV color-space augment_hsv(img, hgain=hyp['hsv_h'], sgain=hyp['hsv_s'], vgain=hyp['hsv_v']) # Flip up-down if random.random() < hyp['flipud']: img = np.flipud(img) if nl: labels[:, 2] = 1 - labels[:, 2] # Flip left-right if random.random() < hyp['fliplr']: img = np.fliplr(img) if nl: labels[:, 1] = 1 - labels[:, 1] # Cutouts # labels = cutout(img, labels, p=0.5) # nl = len(labels) # update after cutout labels_out = torch.zeros((nl, 6)) if nl: labels_out[:, 1:] = torch.from_numpy(labels) # Convert img = img.transpose((2, 0, 1))[::-1] # HWC to CHW, BGR to RGB img = np.ascontiguousarray(img) return torch.from_numpy(img), labels_out, self.im_files[index], shapes def load_image(self, i): # Loads 1 image from dataset index 'i', returns (im, original hw, resized hw) im, f, fn = self.ims[i], self.im_files[i], self.npy_files[i], if im is None: # not cached in RAM if fn.exists(): # load npy im = np.load(fn) else: # read image im = cv2.imread(f) # BGR assert im is not None, f'Image Not Found {f}' h0, w0 = im.shape[:2] # orig hw r = self.img_size / max(h0, w0) # ratio if r != 1: # if sizes are not equal im = cv2.resize(im, (int(w0 * r), int(h0 * r)), interpolation=cv2.INTER_LINEAR if (self.augment or r > 1) else cv2.INTER_AREA) return im, (h0, w0), im.shape[:2] # im, hw_original, hw_resized else: return self.ims[i], self.im_hw0[i], self.im_hw[i] # im, hw_original, hw_resized def cache_images_to_disk(self, i): # Saves an image as an *.npy file for faster loading f = self.npy_files[i] if not f.exists(): np.save(f.as_posix(), cv2.imread(self.im_files[i])) def load_mosaic(self, index): # YOLOv5 4-mosaic loader. Loads 1 image + 3 random images into a 4-image mosaic labels4, segments4 = [], [] s = self.img_size yc, xc = (int(random.uniform(-x, 2 * s + x)) for x in self.mosaic_border) # mosaic center x, y indices = [index] + random.choices(self.indices, k=3) # 3 additional image indices random.shuffle(indices) for i, index in enumerate(indices): # Load image img, _, (h, w) = self.load_image(index) # place img in img4 if i == 0: # top left img4 = np.full((s * 2, s * 2, img.shape[2]), 114, dtype=np.uint8) # base image with 4 tiles x1a, y1a, x2a, y2a = max(xc - w, 0), max(yc - h, 0), xc, yc # xmin, ymin, xmax, ymax (large image) x1b, y1b, x2b, y2b = w - (x2a - x1a), h - (y2a - y1a), w, h # xmin, ymin, xmax, ymax (small image) elif i == 1: # top right x1a, y1a, x2a, y2a = xc, max(yc - h, 0), min(xc + w, s * 2), yc x1b, y1b, x2b, y2b = 0, h - (y2a - y1a), min(w, x2a - x1a), h elif i == 2: # bottom left x1a, y1a, x2a, y2a = max(xc - w, 0), yc, xc, min(s * 2, yc + h) x1b, y1b, x2b, y2b = w - (x2a - x1a), 0, w, min(y2a - y1a, h) elif i == 3: # bottom right x1a, y1a, x2a, y2a = xc, yc, min(xc + w, s * 2), min(s * 2, yc + h) x1b, y1b, x2b, y2b = 0, 0, min(w, x2a - x1a), min(y2a - y1a, h) img4[y1a:y2a, x1a:x2a] = img[y1b:y2b, x1b:x2b] # img4[ymin:ymax, xmin:xmax] padw = x1a - x1b padh = y1a - y1b # Labels labels, segments = self.labels[index].copy(), self.segments[index].copy() if labels.size: labels[:, 1:] = xywhn2xyxy(labels[:, 1:], w, h, padw, padh) # normalized xywh to pixel xyxy format segments = [xyn2xy(x, w, h, padw, padh) for x in segments] labels4.append(labels) segments4.extend(segments) # Concat/clip labels labels4 = np.concatenate(labels4, 0) for x in (labels4[:, 1:], *segments4): np.clip(x, 0, 2 * s, out=x) # clip when using random_perspective() # img4, labels4 = replicate(img4, labels4) # replicate # Augment img4, labels4, segments4 = copy_paste(img4, labels4, segments4, p=self.hyp['copy_paste']) img4, labels4 = random_perspective(img4, labels4, segments4, degrees=self.hyp['degrees'], translate=self.hyp['translate'], scale=self.hyp['scale'], shear=self.hyp['shear'], perspective=self.hyp['perspective'], border=self.mosaic_border) # border to remove return img4, labels4 def load_mosaic9(self, index): # YOLOv5 9-mosaic loader. Loads 1 image + 8 random images into a 9-image mosaic labels9, segments9 = [], [] s = self.img_size indices = [index] + random.choices(self.indices, k=8) # 8 additional image indices random.shuffle(indices) hp, wp = -1, -1 # height, width previous for i, index in enumerate(indices): # Load image img, _, (h, w) = self.load_image(index) # place img in img9 if i == 0: # center img9 = np.full((s * 3, s * 3, img.shape[2]), 114, dtype=np.uint8) # base image with 4 tiles h0, w0 = h, w c = s, s, s + w, s + h # xmin, ymin, xmax, ymax (base) coordinates elif i == 1: # top c = s, s - h, s + w, s elif i == 2: # top right c = s + wp, s - h, s + wp + w, s elif i == 3: # right c = s + w0, s, s + w0 + w, s + h elif i == 4: # bottom right c = s + w0, s + hp, s + w0 + w, s + hp + h elif i == 5: # bottom c = s + w0 - w, s + h0, s + w0, s + h0 + h elif i == 6: # bottom left c = s + w0 - wp - w, s + h0, s + w0 - wp, s + h0 + h elif i == 7: # left c = s - w, s + h0 - h, s, s + h0 elif i == 8: # top left c = s - w, s + h0 - hp - h, s, s + h0 - hp padx, pady = c[:2] x1, y1, x2, y2 = (max(x, 0) for x in c) # allocate coords # Labels labels, segments = self.labels[index].copy(), self.segments[index].copy() if labels.size: labels[:, 1:] = xywhn2xyxy(labels[:, 1:], w, h, padx, pady) # normalized xywh to pixel xyxy format segments = [xyn2xy(x, w, h, padx, pady) for x in segments] labels9.append(labels) segments9.extend(segments) # Image img9[y1:y2, x1:x2] = img[y1 - pady:, x1 - padx:] # img9[ymin:ymax, xmin:xmax] hp, wp = h, w # height, width previous # Offset yc, xc = (int(random.uniform(0, s)) for _ in self.mosaic_border) # mosaic center x, y img9 = img9[yc:yc + 2 * s, xc:xc + 2 * s] # Concat/clip labels labels9 = np.concatenate(labels9, 0) labels9[:, [1, 3]] -= xc labels9[:, [2, 4]] -= yc c = np.array([xc, yc]) # centers segments9 = [x - c for x in segments9] for x in (labels9[:, 1:], *segments9): np.clip(x, 0, 2 * s, out=x) # clip when using random_perspective() # img9, labels9 = replicate(img9, labels9) # replicate # Augment img9, labels9 = random_perspective(img9, labels9, segments9, degrees=self.hyp['degrees'], translate=self.hyp['translate'], scale=self.hyp['scale'], shear=self.hyp['shear'], perspective=self.hyp['perspective'], border=self.mosaic_border) # border to remove return img9, labels9 @staticmethod def collate_fn(batch): im, label, path, shapes = zip(*batch) # transposed for i, lb in enumerate(label): lb[:, 0] = i # add target image index for build_targets() return torch.stack(im, 0), torch.cat(label, 0), path, shapes @staticmethod def collate_fn4(batch): img, label, path, shapes = zip(*batch) # transposed n = len(shapes) // 4 im4, label4, path4, shapes4 = [], [], path[:n], shapes[:n] ho = torch.tensor([[0.0, 0, 0, 1, 0, 0]]) wo = torch.tensor([[0.0, 0, 1, 0, 0, 0]]) s = torch.tensor([[1, 1, 0.5, 0.5, 0.5, 0.5]]) # scale for i in range(n): # zidane torch.zeros(16,3,720,1280) # BCHW i *= 4 if random.random() < 0.5: im = F.interpolate(img[i].unsqueeze(0).float(), scale_factor=2.0, mode='bilinear', align_corners=False)[0].type(img[i].type()) lb = label[i] else: im = torch.cat((torch.cat((img[i], img[i + 1]), 1), torch.cat((img[i + 2], img[i + 3]), 1)), 2) lb = torch.cat((label[i], label[i + 1] + ho, label[i + 2] + wo, label[i + 3] + ho + wo), 0) * s im4.append(im) label4.append(lb) for i, lb in enumerate(label4): lb[:, 0] = i # add target image index for build_targets() return torch.stack(im4, 0), torch.cat(label4, 0), path4, shapes4 # Ancillary functions -------------------------------------------------------------------------------------------------- def create_folder(path='./new'): # Create folder if os.path.exists(path): shutil.rmtree(path) # delete output folder os.makedirs(path) # make new output folder def flatten_recursive(path=DATASETS_DIR / 'coco128'): # Flatten a recursive directory by bringing all files to top level new_path = Path(str(path) + '_flat') create_folder(new_path) for file in tqdm(glob.glob(str(Path(path)) + '/**/*.*', recursive=True)): shutil.copyfile(file, new_path / Path(file).name) def extract_boxes(path=DATASETS_DIR / 'coco128'): # from utils.datasets import *; extract_boxes() # Convert detection dataset into classification dataset, with one directory per class path = Path(path) # images dir shutil.rmtree(path / 'classifier') if (path / 'classifier').is_dir() else None # remove existing files = list(path.rglob('*.*')) n = len(files) # number of files for im_file in tqdm(files, total=n): if im_file.suffix[1:] in IMG_FORMATS: # image im = cv2.imread(str(im_file))[..., ::-1] # BGR to RGB h, w = im.shape[:2] # labels lb_file = Path(img2label_paths([str(im_file)])[0]) if Path(lb_file).exists(): with open(lb_file) as f: lb = np.array([x.split() for x in f.read().strip().splitlines()], dtype=np.float32) # labels for j, x in enumerate(lb): c = int(x[0]) # class f = (path / 'classifier') / f'{c}' / f'{path.stem}_{im_file.stem}_{j}.jpg' # new filename if not f.parent.is_dir(): f.parent.mkdir(parents=True) b = x[1:] * [w, h, w, h] # box # b[2:] = b[2:].max() # rectangle to square b[2:] = b[2:] * 1.2 + 3 # pad b = xywh2xyxy(b.reshape(-1, 4)).ravel().astype(np.int32) b[[0, 2]] = np.clip(b[[0, 2]], 0, w) # clip boxes outside of image b[[1, 3]] = np.clip(b[[1, 3]], 0, h) assert cv2.imwrite(str(f), im[b[1]:b[3], b[0]:b[2]]), f'box failure in {f}' def autosplit(path=DATASETS_DIR / 'coco128/images', weights=(0.9, 0.1, 0.0), annotated_only=False): """ Autosplit a dataset into train/val/test splits and save path/autosplit_*.txt files Usage: from utils.datasets import *; autosplit() Arguments path: Path to images directory weights: Train, val, test weights (list, tuple) annotated_only: Only use images with an annotated txt file """ path = Path(path) # images dir files = sorted(x for x in path.rglob('*.*') if x.suffix[1:].lower() in IMG_FORMATS) # image files only n = len(files) # number of files random.seed(0) # for reproducibility indices = random.choices([0, 1, 2], weights=weights, k=n) # assign each image to a split txt = ['autosplit_train.txt', 'autosplit_val.txt', 'autosplit_test.txt'] # 3 txt files [(path.parent / x).unlink(missing_ok=True) for x in txt] # remove existing print(f'Autosplitting images from {path}' + ', using *.txt labeled images only' * annotated_only) for i, img in tqdm(zip(indices, files), total=n): if not annotated_only or Path(img2label_paths([str(img)])[0]).exists(): # check label with open(path.parent / txt[i], 'a') as f: f.write('./' + img.relative_to(path.parent).as_posix() + '\n') # add image to txt file def verify_image_label(args): # Verify one image-label pair im_file, lb_file, prefix = args nm, nf, ne, nc, msg, segments = 0, 0, 0, 0, '', [] # number (missing, found, empty, corrupt), message, segments try: # verify images im = Image.open(im_file) im.verify() # PIL verify shape = exif_size(im) # image size assert (shape[0] > 9) & (shape[1] > 9), f'image size {shape} <10 pixels' assert im.format.lower() in IMG_FORMATS, f'invalid image format {im.format}' if im.format.lower() in ('jpg', 'jpeg'): with open(im_file, 'rb') as f: f.seek(-2, 2) if f.read() != b'\xff\xd9': # corrupt JPEG ImageOps.exif_transpose(Image.open(im_file)).save(im_file, 'JPEG', subsampling=0, quality=100) msg = f'{prefix}WARNING: {im_file}: corrupt JPEG restored and saved' # verify labels if os.path.isfile(lb_file): nf = 1 # label found with open(lb_file) as f: lb = [x.split() for x in f.read().strip().splitlines() if len(x)] if any(len(x) > 6 for x in lb): # is segment classes = np.array([x[0] for x in lb], dtype=np.float32) segments = [np.array(x[1:], dtype=np.float32).reshape(-1, 2) for x in lb] # (cls, xy1...) lb = np.concatenate((classes.reshape(-1, 1), segments2boxes(segments)), 1) # (cls, xywh) lb = np.array(lb, dtype=np.float32) nl = len(lb) if nl: assert lb.shape[1] == 5, f'labels require 5 columns, {lb.shape[1]} columns detected' assert (lb >= 0).all(), f'negative label values {lb[lb < 0]}' assert (lb[:, 1:] <= 1).all(), f'non-normalized or out of bounds coordinates {lb[:, 1:][lb[:, 1:] > 1]}' _, i = np.unique(lb, axis=0, return_index=True) if len(i) < nl: # duplicate row check lb = lb[i] # remove duplicates if segments: segments = segments[i] msg = f'{prefix}WARNING: {im_file}: {nl - len(i)} duplicate labels removed' else: ne = 1 # label empty lb = np.zeros((0, 5), dtype=np.float32) else: nm = 1 # label missing lb = np.zeros((0, 5), dtype=np.float32) return im_file, lb, shape, segments, nm, nf, ne, nc, msg except Exception as e: nc = 1 msg = f'{prefix}WARNING: {im_file}: ignoring corrupt image/label: {e}' return [None, None, None, None, nm, nf, ne, nc, msg] def dataset_stats(path='coco128.yaml', autodownload=False, verbose=False, profile=False, hub=False): """ Return dataset statistics dictionary with images and instances counts per split per class To run in parent directory: export PYTHONPATH="$PWD/yolov5" Usage1: from utils.datasets import *; dataset_stats('coco128.yaml', autodownload=True) Usage2: from utils.datasets import *; dataset_stats('path/to/coco128_with_yaml.zip') Arguments path: Path to data.yaml or data.zip (with data.yaml inside data.zip) autodownload: Attempt to download dataset if not found locally verbose: Print stats dictionary """ def round_labels(labels): # Update labels to integer class and 6 decimal place floats return [[int(c), *(round(x, 4) for x in points)] for c, *points in labels] def unzip(path): # Unzip data.zip TODO: CONSTRAINT: path/to/abc.zip MUST unzip to 'path/to/abc/' if str(path).endswith('.zip'): # path is data.zip assert Path(path).is_file(), f'Error unzipping {path}, file not found' ZipFile(path).extractall(path=path.parent) # unzip dir = path.with_suffix('') # dataset directory == zip name return True, str(dir), next(dir.rglob('*.yaml')) # zipped, data_dir, yaml_path else: # path is data.yaml return False, None, path def hub_ops(f, max_dim=1920): # HUB ops for 1 image 'f': resize and save at reduced quality in /dataset-hub for web/app viewing f_new = im_dir / Path(f).name # dataset-hub image filename try: # use PIL im = Image.open(f) r = max_dim / max(im.height, im.width) # ratio if r < 1.0: # image too large im = im.resize((int(im.width * r), int(im.height * r))) im.save(f_new, 'JPEG', quality=75, optimize=True) # save except Exception as e: # use OpenCV print(f'WARNING: HUB ops PIL failure {f}: {e}') im = cv2.imread(f) im_height, im_width = im.shape[:2] r = max_dim / max(im_height, im_width) # ratio if r < 1.0: # image too large im = cv2.resize(im, (int(im_width * r), int(im_height * r)), interpolation=cv2.INTER_AREA) cv2.imwrite(str(f_new), im) zipped, data_dir, yaml_path = unzip(Path(path)) with open(check_yaml(yaml_path), errors='ignore') as f: data = yaml.safe_load(f) # data dict if zipped: data['path'] = data_dir # TODO: should this be dir.resolve()? check_dataset(data, autodownload) # download dataset if missing hub_dir = Path(data['path'] + ('-hub' if hub else '')) stats = {'nc': data['nc'], 'names': data['names']} # statistics dictionary for split in 'train', 'val', 'test': if data.get(split) is None: stats[split] = None # i.e. no test set continue x = [] dataset = LoadImagesAndLabels(data[split]) # load dataset for label in tqdm(dataset.labels, total=dataset.n, desc='Statistics'): x.append(np.bincount(label[:, 0].astype(int), minlength=data['nc'])) x = np.array(x) # shape(128x80) stats[split] = { 'instance_stats': { 'total': int(x.sum()), 'per_class': x.sum(0).tolist()}, 'image_stats': { 'total': dataset.n, 'unlabelled': int(np.all(x == 0, 1).sum()), 'per_class': (x > 0).sum(0).tolist()}, 'labels': [{ str(Path(k).name): round_labels(v.tolist())} for k, v in zip(dataset.im_files, dataset.labels)]} if hub: im_dir = hub_dir / 'images' im_dir.mkdir(parents=True, exist_ok=True) for _ in tqdm(ThreadPool(NUM_THREADS).imap(hub_ops, dataset.im_files), total=dataset.n, desc='HUB Ops'): pass # Profile stats_path = hub_dir / 'stats.json' if profile: for _ in range(1): file = stats_path.with_suffix('.npy') t1 = time.time() np.save(file, stats) t2 = time.time() x = np.load(file, allow_pickle=True) print(f'stats.npy times: {time.time() - t2:.3f}s read, {t2 - t1:.3f}s write') file = stats_path.with_suffix('.json') t1 = time.time() with open(file, 'w') as f: json.dump(stats, f) # save stats *.json t2 = time.time() with open(file) as f: x = json.load(f) # load hyps dict print(f'stats.json times: {time.time() - t2:.3f}s read, {t2 - t1:.3f}s write') # Save, print and return if hub: print(f'Saving {stats_path.resolve()}...') with open(stats_path, 'w') as f: json.dump(stats, f) # save stats.json if verbose: print(json.dumps(stats, indent=2, sort_keys=False)) return stats ================================================ FILE: RStask/ObjectDetection/utils/docker/Dockerfile ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Start FROM Nvidia PyTorch image https://ngc.nvidia.com/catalog/containers/nvidia:pytorch FROM nvcr.io/nvidia/pytorch:21.10-py3 # Install linux packages RUN apt update && apt install -y zip htop screen libgl1-mesa-glx # Install python dependencies COPY requirements.txt . RUN python -m pip install --upgrade pip RUN pip uninstall -y torch torchvision torchtext RUN pip install --no-cache -r requirements.txt albumentations wandb gsutil notebook \ torch==1.11.0+cu113 torchvision==0.12.0+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html # RUN pip install --no-cache -U torch torchvision # Create working directory RUN mkdir -p /usr/src/app WORKDIR /usr/src/app # Copy contents COPY . /usr/src/app RUN git clone https://github.com/ultralytics/yolov5 /usr/src/yolov5 # Downloads to user config dir ADD https://ultralytics.com/assets/Arial.ttf https://ultralytics.com/assets/Arial.Unicode.ttf /root/.config/Ultralytics/ # Set environment variables ENV OMP_NUM_THREADS=8 # Usage Examples ------------------------------------------------------------------------------------------------------- # Build and Push # t=ultralytics/yolov5:latest && sudo docker build -f utils/docker/Dockerfile -t $t . && sudo docker push $t # Pull and Run # t=ultralytics/yolov5:latest && sudo docker pull $t && sudo docker run -it --ipc=host --gpus all $t # Pull and Run with local directory access # t=ultralytics/yolov5:latest && sudo docker pull $t && sudo docker run -it --ipc=host --gpus all -v "$(pwd)"/datasets:/usr/src/datasets $t # Kill all # sudo docker kill $(sudo docker ps -q) # Kill all image-based # sudo docker kill $(sudo docker ps -qa --filter ancestor=ultralytics/yolov5:latest) # Bash into running container # sudo docker exec -it 5a9b5863d93d bash # Bash into stopped container # id=$(sudo docker ps -qa) && sudo docker start $id && sudo docker exec -it $id bash # Clean up # docker system prune -a --volumes # Update Ubuntu drivers # https://www.maketecheasier.com/install-nvidia-drivers-ubuntu/ # DDP test # python -m torch.distributed.run --nproc_per_node 2 --master_port 1 train.py --epochs 3 # GCP VM from Image # docker.io/ultralytics/yolov5:latest ================================================ FILE: RStask/ObjectDetection/utils/docker/Dockerfile-cpu ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license # Start FROM Ubuntu image https://hub.docker.com/_/ubuntu FROM ubuntu:latest # Install linux packages RUN apt update RUN DEBIAN_FRONTEND=noninteractive TZ=Etc/UTC apt install -y tzdata RUN apt install -y python3-pip git zip curl htop screen libgl1-mesa-glx libglib2.0-0 RUN alias python=python3 # Install python dependencies COPY requirements.txt . RUN python3 -m pip install --upgrade pip RUN pip install --no-cache -r requirements.txt albumentations gsutil notebook \ coremltools onnx onnx-simplifier onnxruntime openvino-dev tensorflow-cpu tensorflowjs \ torch==1.11.0+cpu torchvision==0.12.0+cpu -f https://download.pytorch.org/whl/cpu/torch_stable.html # Create working directory RUN mkdir -p /usr/src/app WORKDIR /usr/src/app # Copy contents COPY . /usr/src/app RUN git clone https://github.com/ultralytics/yolov5 /usr/src/yolov5 # Downloads to user config dir ADD https://ultralytics.com/assets/Arial.ttf https://ultralytics.com/assets/Arial.Unicode.ttf /root/.config/Ultralytics/ # Usage Examples ------------------------------------------------------------------------------------------------------- # Build and Push # t=ultralytics/yolov5:latest-cpu && sudo docker build -f utils/docker/Dockerfile-cpu -t $t . && sudo docker push $t # Pull and Run # t=ultralytics/yolov5:latest-cpu && sudo docker pull $t && sudo docker run -it --ipc=host -v "$(pwd)"/datasets:/usr/src/datasets $t ================================================ FILE: RStask/ObjectDetection/utils/downloads.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Download utils """ import logging import os import platform import subprocess import time import urllib from pathlib import Path from zipfile import ZipFile import requests import torch def gsutil_getsize(url=''): # gs://bucket/file size https://cloud.google.com/storage/docs/gsutil/commands/du s = subprocess.check_output(f'gsutil du {url}', shell=True).decode('utf-8') return eval(s.split(' ')[0]) if len(s) else 0 # bytes def safe_download(file, url, url2=None, min_bytes=1E0, error_msg=''): # Attempts to download file from url or url2, checks and removes incomplete downloads < min_bytes from ObjectDetection.utils.general import LOGGER file = Path(file) assert_msg = f"Downloaded file '{file}' does not exist or size is < min_bytes={min_bytes}" try: # url1 LOGGER.info(f'Downloading {url} to {file}...') torch.hub.download_url_to_file(url, str(file), progress=LOGGER.level <= logging.INFO) assert file.exists() and file.stat().st_size > min_bytes, assert_msg # check except Exception as e: # url2 file.unlink(missing_ok=True) # remove partial downloads LOGGER.info(f'ERROR: {e}\nRe-attempting {url2 or url} to {file}...') os.system(f"curl -L '{url2 or url}' -o '{file}' --retry 3 -C -") # curl download, retry and resume on fail finally: if not file.exists() or file.stat().st_size < min_bytes: # check file.unlink(missing_ok=True) # remove partial downloads LOGGER.info(f"ERROR: {assert_msg}\n{error_msg}") LOGGER.info('') def attempt_download(file, repo='ultralytics/yolov5'): # from utils.downloads import *; attempt_download() # Attempt file download if does not exist from RStask.ObjectDetection.utils.general import LOGGER file = Path(str(file).strip().replace("'", '')) if not file.exists(): # URL specified name = Path(urllib.parse.unquote(str(file))).name # decode '%2F' to '/' etc. if str(file).startswith(('http:/', 'https:/')): # download url = str(file).replace(':/', '://') # Pathlib turns :// -> :/ file = name.split('?')[0] # parse authentication https://url.com/file.txt?auth... if Path(file).is_file(): LOGGER.info(f'Found {url} locally at {file}') # file already exists else: safe_download(file=file, url=url, min_bytes=1E5) return file # GitHub assets file.parent.mkdir(parents=True, exist_ok=True) # make parent dir (if required) try: response = requests.get(f'https://api.github.com/repos/{repo}/releases/latest').json() # github api assets = [x['name'] for x in response['assets']] # release assets, i.e. ['yolov5s.pt', 'yolov5m.pt', ...] tag = response['tag_name'] # i.e. 'v1.0' except Exception: # fallback plan assets = [ 'yolov5n.pt', 'yolov5s.pt', 'yolov5m.pt', 'yolov5l.pt', 'yolov5x.pt', 'yolov5n6.pt', 'yolov5s6.pt', 'yolov5m6.pt', 'yolov5l6.pt', 'yolov5x6.pt'] try: tag = subprocess.check_output('git tag', shell=True, stderr=subprocess.STDOUT).decode().split()[-1] except Exception: tag = 'v6.1' # current release if name in assets: url3 = 'https://drive.google.com/drive/folders/1EFQTEUeXWSFww0luse2jB9M1QNZQGwNl' # backup gdrive mirror safe_download( file, url=f'https://github.com/{repo}/releases/download/{tag}/{name}', url2=f'https://storage.googleapis.com/{repo}/{tag}/{name}', # backup url (optional) min_bytes=1E5, error_msg=f'{file} missing, try downloading from https://github.com/{repo}/releases/{tag} or {url3}') return str(file) def gdrive_download(id='16TiPfZj7htmTyhntwcZyEEAejOUxuT6m', file='tmp.zip'): # Downloads a file from Google Drive. from yolov5.utils.downloads import *; gdrive_download() t = time.time() file = Path(file) cookie = Path('cookie') # gdrive cookie print(f'Downloading https://drive.google.com/uc?export=download&id={id} as {file}... ', end='') file.unlink(missing_ok=True) # remove existing file cookie.unlink(missing_ok=True) # remove existing cookie # Attempt file download out = "NUL" if platform.system() == "Windows" else "/dev/null" os.system(f'curl -c ./cookie -s -L "drive.google.com/uc?export=download&id={id}" > {out}') if os.path.exists('cookie'): # large file s = f'curl -Lb ./cookie "drive.google.com/uc?export=download&confirm={get_token()}&id={id}" -o {file}' else: # small file s = f'curl -s -L -o {file} "drive.google.com/uc?export=download&id={id}"' r = os.system(s) # execute, capture return cookie.unlink(missing_ok=True) # remove existing cookie # Error check if r != 0: file.unlink(missing_ok=True) # remove partial print('Download error ') # raise Exception('Download error') return r # Unzip if archive if file.suffix == '.zip': print('unzipping... ', end='') ZipFile(file).extractall(path=file.parent) # unzip file.unlink() # remove zip print(f'Done ({time.time() - t:.1f}s)') return r def get_token(cookie="./cookie"): with open(cookie) as f: for line in f: if "download" in line: return line.split()[-1] return "" # Google utils: https://cloud.google.com/storage/docs/reference/libraries ---------------------------------------------- # # # def upload_blob(bucket_name, source_file_name, destination_blob_name): # # Uploads a file to a bucket # # https://cloud.google.com/storage/docs/uploading-objects#storage-upload-object-python # # storage_client = storage.Client() # bucket = storage_client.get_bucket(bucket_name) # blob = bucket.blob(destination_blob_name) # # blob.upload_from_filename(source_file_name) # # print('File {} uploaded to {}.'.format( # source_file_name, # destination_blob_name)) # # # def download_blob(bucket_name, source_blob_name, destination_file_name): # # Uploads a blob from a bucket # storage_client = storage.Client() # bucket = storage_client.get_bucket(bucket_name) # blob = bucket.blob(source_blob_name) # # blob.download_to_filename(destination_file_name) # # print('Blob {} downloaded to {}.'.format( # source_blob_name, # destination_file_name)) ================================================ FILE: RStask/ObjectDetection/utils/flask_rest_api/README.md ================================================ # Flask REST API [REST](https://en.wikipedia.org/wiki/Representational_state_transfer) [API](https://en.wikipedia.org/wiki/API)s are commonly used to expose Machine Learning (ML) models to other services. This folder contains an example REST API created using Flask to expose the YOLOv5s model from [PyTorch Hub](https://pytorch.org/hub/ultralytics_yolov5/). ## Requirements [Flask](https://palletsprojects.com/p/flask/) is required. Install with: ```shell $ pip install Flask ``` ## Run After Flask installation run: ```shell $ python3 restapi.py --port 5000 ``` Then use [curl](https://curl.se/) to perform a request: ```shell $ curl -X POST -F image=@zidane.jpg 'http://localhost:5000/v1/object-detection/yolov5s' ``` The model inference results are returned as a JSON response: ```json [ { "class": 0, "confidence": 0.8900438547, "height": 0.9318675399, "name": "person", "width": 0.3264600933, "xcenter": 0.7438579798, "ycenter": 0.5207948685 }, { "class": 0, "confidence": 0.8440024257, "height": 0.7155083418, "name": "person", "width": 0.6546785235, "xcenter": 0.427829951, "ycenter": 0.6334488392 }, { "class": 27, "confidence": 0.3771208823, "height": 0.3902671337, "name": "tie", "width": 0.0696444362, "xcenter": 0.3675483763, "ycenter": 0.7991207838 }, { "class": 27, "confidence": 0.3527112305, "height": 0.1540903747, "name": "tie", "width": 0.0336618312, "xcenter": 0.7814827561, "ycenter": 0.5065554976 } ] ``` An example python script to perform inference using [requests](https://docs.python-requests.org/en/master/) is given in `example_request.py` ================================================ FILE: RStask/ObjectDetection/utils/flask_rest_api/example_request.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Perform test request """ import pprint import requests DETECTION_URL = "http://localhost:5000/v1/object-detection/yolov5s" IMAGE = "zidane.jpg" # Read image with open(IMAGE, "rb") as f: image_data = f.read() response = requests.post(DETECTION_URL, files={"image": image_data}).json() pprint.pprint(response) ================================================ FILE: RStask/ObjectDetection/utils/flask_rest_api/restapi.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Run a Flask REST API exposing a YOLOv5s model """ import argparse import io import torch from flask import Flask, request from PIL import Image app = Flask(__name__) DETECTION_URL = "/v1/object-detection/yolov5s" @app.route(DETECTION_URL, methods=["POST"]) def predict(): if not request.method == "POST": return if request.files.get("image"): # Method 1 # with request.files["image"] as f: # im = Image.open(io.BytesIO(f.read())) # Method 2 im_file = request.files["image"] im_bytes = im_file.read() im = Image.open(io.BytesIO(im_bytes)) results = model(im, size=640) # reduce size=320 for faster inference return results.pandas().xyxy[0].to_json(orient="records") if __name__ == "__main__": parser = argparse.ArgumentParser(description="Flask API exposing YOLOv5 model") parser.add_argument("--port", default=5000, type=int, help="port number") opt = parser.parse_args() # Fix known issue urllib.error.HTTPError 403: rate limit exceeded https://github.com/ultralytics/yolov5/pull/7210 torch.hub._validate_not_a_forked_repo = lambda a, b, c: True model = torch.hub.load("ultralytics/yolov5", "yolov5s", force_reload=True) # force_reload to recache app.run(host="0.0.0.0", port=opt.port) # debug=True causes Restarting with stat ================================================ FILE: RStask/ObjectDetection/utils/general.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ General utils """ import contextlib import glob import inspect import logging import math import os import platform import random import re import shutil import signal import time import urllib from datetime import datetime from itertools import repeat from multiprocessing.pool import ThreadPool from pathlib import Path from subprocess import check_output from typing import Optional from zipfile import ZipFile import cv2 import numpy as np import pandas as pd import pkg_resources as pkg import torch import torchvision import yaml from RStask.ObjectDetection.utils.downloads import gsutil_getsize from RStask.ObjectDetection.utils.metrics import box_iou, fitness # Settings FILE = Path(__file__).resolve() ROOT = FILE.parents[1] # YOLOv5 root directory DATASETS_DIR = ROOT.parent / 'datasets' # YOLOv5 datasets directory NUM_THREADS = min(8, max(1, os.cpu_count() - 1)) # number of YOLOv5 multiprocessing threads AUTOINSTALL = str(os.getenv('YOLOv5_AUTOINSTALL', True)).lower() == 'true' # global auto-install mode VERBOSE = str(os.getenv('YOLOv5_VERBOSE', True)).lower() == 'true' # global verbose mode FONT = 'Arial.ttf' # https://ultralytics.com/assets/Arial.ttf torch.set_printoptions(linewidth=320, precision=5, profile='long') np.set_printoptions(linewidth=320, formatter={'float_kind': '{:11.5g}'.format}) # format short g, %precision=5 pd.options.display.max_columns = 10 cv2.setNumThreads(0) # prevent OpenCV from multithreading (incompatible with PyTorch DataLoader) os.environ['NUMEXPR_MAX_THREADS'] = str(NUM_THREADS) # NumExpr max threads os.environ['OMP_NUM_THREADS'] = str(NUM_THREADS) # OpenMP max threads (PyTorch and SciPy) def is_kaggle(): # Is environment a Kaggle Notebook? try: assert os.environ.get('PWD') == '/kaggle/working' assert os.environ.get('KAGGLE_URL_BASE') == 'https://www.kaggle.com' return True except AssertionError: return False def is_writeable(dir, test=False): # Return True if directory has write permissions, test opening a file with write permissions if test=True if test: # method 1 file = Path(dir) / 'tmp.txt' try: with open(file, 'w'): # open file with write permissions pass file.unlink() # remove file return True except OSError: return False else: # method 2 return os.access(dir, os.R_OK) # possible issues on Windows def set_logging(name=None, verbose=VERBOSE): # Sets level and returns logger if is_kaggle(): for h in logging.root.handlers: logging.root.removeHandler(h) # remove all handlers associated with the root logger object rank = int(os.getenv('RANK', -1)) # rank in world for Multi-GPU trainings level = logging.INFO if (verbose and rank in (-1, 0)) else logging.WARNING log = logging.getLogger(name) log.setLevel(level) handler = logging.StreamHandler() handler.setFormatter(logging.Formatter("%(message)s")) handler.setLevel(level) log.addHandler(handler) set_logging() # run before defining LOGGER LOGGER = logging.getLogger("yolov5") # define globally (used in train.py, val.py, detect.py, etc.) def user_config_dir(dir='Ultralytics', env_var='YOLOV5_CONFIG_DIR'): # Return path of user configuration directory. Prefer environment variable if exists. Make dir if required. env = os.getenv(env_var) if env: path = Path(env) # use environment variable else: cfg = {'Windows': 'AppData/Roaming', 'Linux': '.config', 'Darwin': 'Library/Application Support'} # 3 OS dirs path = Path.home() / cfg.get(platform.system(), '') # OS-specific config dir path = (path if is_writeable(path) else Path('/tmp')) / dir # GCP and AWS lambda fix, only /tmp is writeable path.mkdir(exist_ok=True) # make if required return path CONFIG_DIR = user_config_dir() # Ultralytics settings dir class Profile(contextlib.ContextDecorator): # Usage: @Profile() decorator or 'with Profile():' context manager def __enter__(self): self.start = time.time() def __exit__(self, type, value, traceback): print(f'Profile results: {time.time() - self.start:.5f}s') class Timeout(contextlib.ContextDecorator): # Usage: @Timeout(seconds) decorator or 'with Timeout(seconds):' context manager def __init__(self, seconds, *, timeout_msg='', suppress_timeout_errors=True): self.seconds = int(seconds) self.timeout_message = timeout_msg self.suppress = bool(suppress_timeout_errors) def _timeout_handler(self, signum, frame): raise TimeoutError(self.timeout_message) def __enter__(self): if platform.system() != 'Windows': # not supported on Windows signal.signal(signal.SIGALRM, self._timeout_handler) # Set handler for SIGALRM signal.alarm(self.seconds) # start countdown for SIGALRM to be raised def __exit__(self, exc_type, exc_val, exc_tb): if platform.system() != 'Windows': signal.alarm(0) # Cancel SIGALRM if it's scheduled if self.suppress and exc_type is TimeoutError: # Suppress TimeoutError return True class WorkingDirectory(contextlib.ContextDecorator): # Usage: @WorkingDirectory(dir) decorator or 'with WorkingDirectory(dir):' context manager def __init__(self, new_dir): self.dir = new_dir # new dir self.cwd = Path.cwd().resolve() # current dir def __enter__(self): os.chdir(self.dir) def __exit__(self, exc_type, exc_val, exc_tb): os.chdir(self.cwd) def try_except(func): # try-except function. Usage: @try_except decorator def handler(*args, **kwargs): try: func(*args, **kwargs) except Exception as e: print(e) return handler def methods(instance): # Get class/instance methods return [f for f in dir(instance) if callable(getattr(instance, f)) and not f.startswith("__")] def print_args(args: Optional[dict] = None, show_file=True, show_fcn=False): # Print function arguments (optional args dict) x = inspect.currentframe().f_back # previous frame file, _, fcn, _, _ = inspect.getframeinfo(x) if args is None: # get args automatically args, _, _, frm = inspect.getargvalues(x) args = {k: v for k, v in frm.items() if k in args} s = (f'{Path(file).stem}: ' if show_file else '') + (f'{fcn}: ' if show_fcn else '') LOGGER.info(colorstr(s) + ', '.join(f'{k}={v}' for k, v in args.items())) def init_seeds(seed=0): # Initialize random number generator (RNG) seeds https://pytorch.org/docs/stable/notes/randomness.html # cudnn seed 0 settings are slower and more reproducible, else faster and less reproducible import torch.backends.cudnn as cudnn random.seed(seed) np.random.seed(seed) torch.manual_seed(seed) cudnn.benchmark, cudnn.deterministic = (False, True) if seed == 0 else (True, False) def intersect_dicts(da, db, exclude=()): # Dictionary intersection of matching keys and shapes, omitting 'exclude' keys, using da values return {k: v for k, v in da.items() if k in db and not any(x in k for x in exclude) and v.shape == db[k].shape} def get_latest_run(search_dir='.'): # Return path to most recent 'last.pt' in /runs (i.e. to --resume from) last_list = glob.glob(f'{search_dir}/**/last*.pt', recursive=True) return max(last_list, key=os.path.getctime) if last_list else '' def is_docker(): # Is environment a Docker container? return Path('/workspace').exists() # or Path('/.dockerenv').exists() def is_colab(): # Is environment a Google Colab instance? try: import google.colab return True except ImportError: return False def is_pip(): # Is file in a pip package? return 'site-packages' in Path(__file__).resolve().parts def is_ascii(s=''): # Is string composed of all ASCII (no UTF) characters? (note str().isascii() introduced in python 3.7) s = str(s) # convert list, tuple, None, etc. to str return len(s.encode().decode('ascii', 'ignore')) == len(s) def is_chinese(s='人工智能'): # Is string composed of any Chinese characters? return True if re.search('[\u4e00-\u9fff]', str(s)) else False def emojis(str=''): # Return platform-dependent emoji-safe version of string return str.encode().decode('ascii', 'ignore') if platform.system() == 'Windows' else str def file_age(path=__file__): # Return days since last file update dt = (datetime.now() - datetime.fromtimestamp(Path(path).stat().st_mtime)) # delta return dt.days # + dt.seconds / 86400 # fractional days def file_update_date(path=__file__): # Return human-readable file modification date, i.e. '2021-3-26' t = datetime.fromtimestamp(Path(path).stat().st_mtime) return f'{t.year}-{t.month}-{t.day}' def file_size(path): # Return file/dir size (MB) mb = 1 << 20 # bytes to MiB (1024 ** 2) path = Path(path) if path.is_file(): return path.stat().st_size / mb elif path.is_dir(): return sum(f.stat().st_size for f in path.glob('**/*') if f.is_file()) / mb else: return 0.0 def check_online(): # Check internet connectivity import socket try: socket.create_connection(("1.1.1.1", 443), 5) # check host accessibility return True except OSError: return False def git_describe(path=ROOT): # path must be a directory # Return human-readable git description, i.e. v5.0-5-g3e25f1e https://git-scm.com/docs/git-describe try: assert (Path(path) / '.git').is_dir() return check_output(f'git -C {path} describe --tags --long --always', shell=True).decode()[:-1] except Exception: return '' @try_except @WorkingDirectory(ROOT) def check_git_status(): # Recommend 'git pull' if code is out of date msg = ', for updates see https://github.com/ultralytics/yolov5' s = colorstr('github: ') # string assert Path('.git').exists(), s + 'skipping check (not a git repository)' + msg assert not is_docker(), s + 'skipping check (Docker image)' + msg assert check_online(), s + 'skipping check (offline)' + msg cmd = 'git fetch && git config --get remote.origin.url' url = check_output(cmd, shell=True, timeout=5).decode().strip().rstrip('.git') # git fetch branch = check_output('git rev-parse --abbrev-ref HEAD', shell=True).decode().strip() # checked out n = int(check_output(f'git rev-list {branch}..origin/master --count', shell=True)) # commits behind if n > 0: s += f"⚠️ YOLOv5 is out of date by {n} commit{'s' * (n > 1)}. Use `git pull` or `git clone {url}` to update." else: s += f'up to date with {url} ✅' LOGGER.info(emojis(s)) # emoji-safe def check_python(minimum='3.7.0'): # Check current python version vs. required python version check_version(platform.python_version(), minimum, name='Python ', hard=True) def check_version(current='0.0.0', minimum='0.0.0', name='version ', pinned=False, hard=False, verbose=False): # Check version vs. required version current, minimum = (pkg.parse_version(x) for x in (current, minimum)) result = (current == minimum) if pinned else (current >= minimum) # bool s = f'{name}{minimum} required by YOLOv5, but {name}{current} is currently installed' # string if hard: assert result, s # assert min requirements met if verbose and not result: LOGGER.warning(s) return result @try_except def check_requirements(requirements=ROOT / 'requirements.txt', exclude=(), install=True, cmds=()): # Check installed dependencies meet requirements (pass *.txt file or list of packages) prefix = colorstr('red', 'bold', 'requirements:') check_python() # check python version if isinstance(requirements, (str, Path)): # requirements.txt file file = Path(requirements) assert file.exists(), f"{prefix} {file.resolve()} not found, check failed." with file.open() as f: requirements = [f'{x.name}{x.specifier}' for x in pkg.parse_requirements(f) if x.name not in exclude] else: # list or tuple of packages requirements = [x for x in requirements if x not in exclude] n = 0 # number of packages updates for i, r in enumerate(requirements): try: pkg.require(r) except Exception: # DistributionNotFound or VersionConflict if requirements not met s = f"{prefix} {r} not found and is required by YOLOv5" if install and AUTOINSTALL: # check environment variable LOGGER.info(f"{s}, attempting auto-update...") try: assert check_online(), f"'pip install {r}' skipped (offline)" LOGGER.info(check_output(f"pip install '{r}' {cmds[i] if cmds else ''}", shell=True).decode()) n += 1 except Exception as e: LOGGER.warning(f'{prefix} {e}') else: LOGGER.info(f'{s}. Please install and rerun your command.') if n: # if packages updated source = file.resolve() if 'file' in locals() else requirements s = f"{prefix} {n} package{'s' * (n > 1)} updated per {source}\n" \ f"{prefix} ⚠️ {colorstr('bold', 'Restart runtime or rerun command for updates to take effect')}\n" LOGGER.info(emojis(s)) def check_img_size(imgsz, s=32, floor=0): # Verify image size is a multiple of stride s in each dimension if isinstance(imgsz, int): # integer i.e. img_size=640 new_size = max(make_divisible(imgsz, int(s)), floor) else: # list i.e. img_size=[640, 480] imgsz = list(imgsz) # convert to list if tuple new_size = [max(make_divisible(x, int(s)), floor) for x in imgsz] if new_size != imgsz: LOGGER.warning(f'WARNING: --img-size {imgsz} must be multiple of max stride {s}, updating to {new_size}') return new_size def check_imshow(): # Check if environment supports image displays try: assert not is_docker(), 'cv2.imshow() is disabled in Docker environments' assert not is_colab(), 'cv2.imshow() is disabled in Google Colab environments' cv2.imshow('test', np.zeros((1, 1, 3))) cv2.waitKey(1) cv2.destroyAllWindows() cv2.waitKey(1) return True except Exception as e: LOGGER.warning(f'WARNING: Environment does not support cv2.imshow() or PIL Image.show() image displays\n{e}') return False def check_suffix(file='yolov5s.pt', suffix=('.pt',), msg=''): # Check file(s) for acceptable suffix if file and suffix: if isinstance(suffix, str): suffix = [suffix] for f in file if isinstance(file, (list, tuple)) else [file]: s = Path(f).suffix.lower() # file suffix if len(s): assert s in suffix, f"{msg}{f} acceptable suffix is {suffix}" def check_yaml(file, suffix=('.yaml', '.yml')): # Search/download YAML file (if necessary) and return path, checking suffix return check_file(file, suffix) def check_file(file, suffix=''): # Search/download file (if necessary) and return path check_suffix(file, suffix) # optional file = str(file) # convert to str() if Path(file).is_file() or file == '': # exists return file elif file.startswith(('http:/', 'https:/')): # download url = str(Path(file)).replace(':/', '://') # Pathlib turns :// -> :/ file = Path(urllib.parse.unquote(file).split('?')[0]).name # '%2F' to '/', split https://url.com/file.txt?auth if Path(file).is_file(): LOGGER.info(f'Found {url} locally at {file}') # file already exists else: LOGGER.info(f'Downloading {url} to {file}...') torch.hub.download_url_to_file(url, file) assert Path(file).exists() and Path(file).stat().st_size > 0, f'File download failed: {url}' # check return file else: # search files = [] for d in 'data', 'models', 'utils': # search directories files.extend(glob.glob(str(ROOT / d / '**' / file), recursive=True)) # find file assert len(files), f'File not found: {file}' # assert file was found assert len(files) == 1, f"Multiple files match '{file}', specify exact path: {files}" # assert unique return files[0] # return file def check_font(font=FONT, progress=False): # Download font to CONFIG_DIR if necessary font = Path(font) file = CONFIG_DIR / font.name if not font.exists() and not file.exists(): url = "https://ultralytics.com/assets/" + font.name LOGGER.info(f'Downloading {url} to {file}...') torch.hub.download_url_to_file(url, str(file), progress=progress) def check_dataset(data, autodownload=True): # Download and/or unzip dataset if not found locally # Usage: https://github.com/ultralytics/yolov5/releases/download/v1.0/coco128_with_yaml.zip # Download (optional) extract_dir = '' # if isinstance(data, (str, Path)) and str(data).endswith('.zip'): # i.e. gs://bucket/dir/coco128.zip # download(data, dir=DATASETS_DIR, unzip=True, delete=False, curl=False, threads=1) # data = next((DATASETS_DIR / Path(data).stem).rglob('*.yaml')) # extract_dir, autodownload = data.parent, False # Read yaml (optional) if isinstance(data, (str, Path)): with open(data, errors='ignore') as f: data = yaml.safe_load(f) # dictionary # Resolve paths path = Path(extract_dir or data.get('path') or '') # optional 'path' default to '.' if not path.is_absolute(): path = (ROOT / path).resolve() for k in 'train', 'val', 'test': if data.get(k): # prepend path data[k] = str(path / data[k]) if isinstance(data[k], str) else [str(path / x) for x in data[k]] # Parse yaml # assert 'nc' in data, "Dataset 'nc' key missing." # if 'names' not in data: # data['names'] = [f'class{i}' for i in range(data['nc'])] # assign class names if missing train, val, test, s = (data.get(x) for x in ('train', 'val', 'test', 'download')) check_font('Arial.ttf' if is_ascii(data['names']) else 'Arial.Unicode.ttf', progress=True) # download fonts return data # dictionary def url2file(url): # Convert URL to filename, i.e. https://url.com/file.txt?auth -> file.txt url = str(Path(url)).replace(':/', '://') # Pathlib turns :// -> :/ file = Path(urllib.parse.unquote(url)).name.split('?')[0] # '%2F' to '/', split https://url.com/file.txt?auth return file def download(url, dir='.', unzip=True, delete=True, curl=False, threads=1, retry=3): # Multi-threaded file download and unzip function, used in data.yaml for autodownload def download_one(url, dir): # Download 1 file success = True f = dir / Path(url).name # filename if Path(url).is_file(): # exists in current path Path(url).rename(f) # move to dir elif not f.exists(): LOGGER.info(f'Downloading {url} to {f}...') for i in range(retry + 1): if curl: s = 'sS' if threads > 1 else '' # silent r = os.system(f"curl -{s}L '{url}' -o '{f}' --retry 9 -C -") # curl download success = r == 0 else: torch.hub.download_url_to_file(url, f, progress=threads == 1) # torch download success = f.is_file() if success: break elif i < retry: LOGGER.warning(f'Download failure, retrying {i + 1}/{retry} {url}...') else: LOGGER.warning(f'Failed to download {url}...') if unzip and success and f.suffix in ('.zip', '.gz'): LOGGER.info(f'Unzipping {f}...') if f.suffix == '.zip': ZipFile(f).extractall(path=dir) # unzip elif f.suffix == '.gz': os.system(f'tar xfz {f} --directory {f.parent}') # unzip if delete: f.unlink() # remove zip dir = Path(dir) dir.mkdir(parents=True, exist_ok=True) # make directory if threads > 1: pool = ThreadPool(threads) pool.imap(lambda x: download_one(*x), zip(url, repeat(dir))) # multi-threaded pool.close() pool.join() else: for u in [url] if isinstance(url, (str, Path)) else url: download_one(u, dir) def make_divisible(x, divisor): # Returns nearest x divisible by divisor if isinstance(divisor, torch.Tensor): divisor = int(divisor.max()) # to int return math.ceil(x / divisor) * divisor def clean_str(s): # Cleans a string by replacing special characters with underscore _ return re.sub(pattern="[|@#!¡·$€%&()=?¿^*;:,¨´><+]", repl="_", string=s) def one_cycle(y1=0.0, y2=1.0, steps=100): # lambda function for sinusoidal ramp from y1 to y2 https://arxiv.org/pdf/1812.01187.pdf return lambda x: ((1 - math.cos(x * math.pi / steps)) / 2) * (y2 - y1) + y1 def colorstr(*input): # Colors a string https://en.wikipedia.org/wiki/ANSI_escape_code, i.e. colorstr('blue', 'hello world') *args, string = input if len(input) > 1 else ('blue', 'bold', input[0]) # color arguments, string colors = { 'black': '\033[30m', # basic colors 'red': '\033[31m', 'green': '\033[32m', 'yellow': '\033[33m', 'blue': '\033[34m', 'magenta': '\033[35m', 'cyan': '\033[36m', 'white': '\033[37m', 'bright_black': '\033[90m', # bright colors 'bright_red': '\033[91m', 'bright_green': '\033[92m', 'bright_yellow': '\033[93m', 'bright_blue': '\033[94m', 'bright_magenta': '\033[95m', 'bright_cyan': '\033[96m', 'bright_white': '\033[97m', 'end': '\033[0m', # misc 'bold': '\033[1m', 'underline': '\033[4m'} return ''.join(colors[x] for x in args) + f'{string}' + colors['end'] def labels_to_class_weights(labels, nc=80): # Get class weights (inverse frequency) from training labels if labels[0] is None: # no labels loaded return torch.Tensor() labels = np.concatenate(labels, 0) # labels.shape = (866643, 5) for COCO classes = labels[:, 0].astype(np.int) # labels = [class xywh] weights = np.bincount(classes, minlength=nc) # occurrences per class # Prepend gridpoint count (for uCE training) # gpi = ((320 / 32 * np.array([1, 2, 4])) ** 2 * 3).sum() # gridpoints per image # weights = np.hstack([gpi * len(labels) - weights.sum() * 9, weights * 9]) ** 0.5 # prepend gridpoints to start weights[weights == 0] = 1 # replace empty bins with 1 weights = 1 / weights # number of targets per class weights /= weights.sum() # normalize return torch.from_numpy(weights) def labels_to_image_weights(labels, nc=80, class_weights=np.ones(80)): # Produces image weights based on class_weights and image contents class_counts = np.array([np.bincount(x[:, 0].astype(np.int), minlength=nc) for x in labels]) image_weights = (class_weights.reshape(1, nc) * class_counts).sum(1) # index = random.choices(range(n), weights=image_weights, k=1) # weight image sample return image_weights def coco80_to_coco91_class(): # converts 80-index (val2014) to 91-index (paper) # https://tech.amikelive.com/node-718/what-object-categories-labels-are-in-coco-dataset/ # a = np.loadtxt('data/coco.names', dtype='str', delimiter='\n') # b = np.loadtxt('data/coco_paper.names', dtype='str', delimiter='\n') # x1 = [list(a[i] == b).index(True) + 1 for i in range(80)] # darknet to coco # x2 = [list(b[i] == a).index(True) if any(b[i] == a) else None for i in range(91)] # coco to darknet x = [ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 27, 28, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 67, 70, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 84, 85, 86, 87, 88, 89, 90] return x def xyxy2xywh(x): # Convert nx4 boxes from [x1, y1, x2, y2] to [x, y, w, h] where xy1=top-left, xy2=bottom-right y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) y[:, 0] = (x[:, 0] + x[:, 2]) / 2 # x center y[:, 1] = (x[:, 1] + x[:, 3]) / 2 # y center y[:, 2] = x[:, 2] - x[:, 0] # width y[:, 3] = x[:, 3] - x[:, 1] # height return y def xywh2xyxy(x): # Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) y[:, 0] = x[:, 0] - x[:, 2] / 2 # top left x y[:, 1] = x[:, 1] - x[:, 3] / 2 # top left y y[:, 2] = x[:, 0] + x[:, 2] / 2 # bottom right x y[:, 3] = x[:, 1] + x[:, 3] / 2 # bottom right y return y def xywhn2xyxy(x, w=640, h=640, padw=0, padh=0): # Convert nx4 boxes from [x, y, w, h] normalized to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) y[:, 0] = w * (x[:, 0] - x[:, 2] / 2) + padw # top left x y[:, 1] = h * (x[:, 1] - x[:, 3] / 2) + padh # top left y y[:, 2] = w * (x[:, 0] + x[:, 2] / 2) + padw # bottom right x y[:, 3] = h * (x[:, 1] + x[:, 3] / 2) + padh # bottom right y return y def xyxy2xywhn(x, w=640, h=640, clip=False, eps=0.0): # Convert nx4 boxes from [x1, y1, x2, y2] to [x, y, w, h] normalized where xy1=top-left, xy2=bottom-right if clip: clip_coords(x, (h - eps, w - eps)) # warning: inplace clip y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) y[:, 0] = ((x[:, 0] + x[:, 2]) / 2) / w # x center y[:, 1] = ((x[:, 1] + x[:, 3]) / 2) / h # y center y[:, 2] = (x[:, 2] - x[:, 0]) / w # width y[:, 3] = (x[:, 3] - x[:, 1]) / h # height return y def xyn2xy(x, w=640, h=640, padw=0, padh=0): # Convert normalized segments into pixel segments, shape (n,2) y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) y[:, 0] = w * x[:, 0] + padw # top left x y[:, 1] = h * x[:, 1] + padh # top left y return y def segment2box(segment, width=640, height=640): # Convert 1 segment label to 1 box label, applying inside-image constraint, i.e. (xy1, xy2, ...) to (xyxy) x, y = segment.T # segment xy inside = (x >= 0) & (y >= 0) & (x <= width) & (y <= height) x, y, = x[inside], y[inside] return np.array([x.min(), y.min(), x.max(), y.max()]) if any(x) else np.zeros((1, 4)) # xyxy def segments2boxes(segments): # Convert segment labels to box labels, i.e. (cls, xy1, xy2, ...) to (cls, xywh) boxes = [] for s in segments: x, y = s.T # segment xy boxes.append([x.min(), y.min(), x.max(), y.max()]) # cls, xyxy return xyxy2xywh(np.array(boxes)) # cls, xywh def resample_segments(segments, n=1000): # Up-sample an (n,2) segment for i, s in enumerate(segments): x = np.linspace(0, len(s) - 1, n) xp = np.arange(len(s)) segments[i] = np.concatenate([np.interp(x, xp, s[:, i]) for i in range(2)]).reshape(2, -1).T # segment xy return segments def scale_coords(img1_shape, coords, img0_shape, ratio_pad=None): # Rescale coords (xyxy) from img1_shape to img0_shape if ratio_pad is None: # calculate from img0_shape gain = min(img1_shape[0] / img0_shape[0], img1_shape[1] / img0_shape[1]) # gain = old / new pad = (img1_shape[1] - img0_shape[1] * gain) / 2, (img1_shape[0] - img0_shape[0] * gain) / 2 # wh padding else: gain = ratio_pad[0][0] pad = ratio_pad[1] coords[:, [0, 2]] -= pad[0] # x padding coords[:, [1, 3]] -= pad[1] # y padding coords[:, :4] /= gain clip_coords(coords, img0_shape) return coords def clip_coords(boxes, shape): # Clip bounding xyxy bounding boxes to image shape (height, width) if isinstance(boxes, torch.Tensor): # faster individually boxes[:, 0].clamp_(0, shape[1]) # x1 boxes[:, 1].clamp_(0, shape[0]) # y1 boxes[:, 2].clamp_(0, shape[1]) # x2 boxes[:, 3].clamp_(0, shape[0]) # y2 else: # np.array (faster grouped) boxes[:, [0, 2]] = boxes[:, [0, 2]].clip(0, shape[1]) # x1, x2 boxes[:, [1, 3]] = boxes[:, [1, 3]].clip(0, shape[0]) # y1, y2 def non_max_suppression(prediction, conf_thres=0.25, iou_thres=0.45, classes=None, agnostic=False, multi_label=False, labels=(), max_det=300): """Non-Maximum Suppression (NMS) on inference results to reject overlapping bounding boxes Returns: list of detections, on (n,6) tensor per image [xyxy, conf, cls] """ bs = prediction.shape[0] # batch size nc = prediction.shape[2] - 5 # number of classes xc = prediction[..., 4] > conf_thres # candidates # Checks assert 0 <= conf_thres <= 1, f'Invalid Confidence threshold {conf_thres}, valid values are between 0.0 and 1.0' assert 0 <= iou_thres <= 1, f'Invalid IoU {iou_thres}, valid values are between 0.0 and 1.0' # Settings # min_wh = 2 # (pixels) minimum box width and height max_wh = 7680 # (pixels) maximum box width and height max_nms = 30000 # maximum number of boxes into torchvision.ops.nms() time_limit = 0.1 + 0.03 * bs # seconds to quit after redundant = True # require redundant detections multi_label &= nc > 1 # multiple labels per box (adds 0.5ms/img) merge = False # use merge-NMS t = time.time() output = [torch.zeros((0, 6), device=prediction.device)] * bs for xi, x in enumerate(prediction): # image index, image inference # Apply constraints # x[((x[..., 2:4] < min_wh) | (x[..., 2:4] > max_wh)).any(1), 4] = 0 # width-height x = x[xc[xi]] # confidence # Cat apriori labels if autolabelling if labels and len(labels[xi]): lb = labels[xi] v = torch.zeros((len(lb), nc + 5), device=x.device) v[:, :4] = lb[:, 1:5] # box v[:, 4] = 1.0 # conf v[range(len(lb)), lb[:, 0].long() + 5] = 1.0 # cls x = torch.cat((x, v), 0) # If none remain process next image if not x.shape[0]: continue # Compute conf x[:, 5:] *= x[:, 4:5] # conf = obj_conf * cls_conf # Box (center x, center y, width, height) to (x1, y1, x2, y2) box = xywh2xyxy(x[:, :4]) # Detections matrix nx6 (xyxy, conf, cls) if multi_label: i, j = (x[:, 5:] > conf_thres).nonzero(as_tuple=False).T x = torch.cat((box[i], x[i, j + 5, None], j[:, None].float()), 1) else: # best class only conf, j = x[:, 5:].max(1, keepdim=True) x = torch.cat((box, conf, j.float()), 1)[conf.view(-1) > conf_thres] # Filter by class if classes is not None: x = x[(x[:, 5:6] == torch.tensor(classes, device=x.device)).any(1)] # Apply finite constraint # if not torch.isfinite(x).all(): # x = x[torch.isfinite(x).all(1)] # Check shape n = x.shape[0] # number of boxes if not n: # no boxes continue elif n > max_nms: # excess boxes x = x[x[:, 4].argsort(descending=True)[:max_nms]] # sort by confidence # Batched NMS c = x[:, 5:6] * (0 if agnostic else max_wh) # classes boxes, scores = x[:, :4] + c, x[:, 4] # boxes (offset by class), scores i = torchvision.ops.nms(boxes, scores, iou_thres) # NMS if i.shape[0] > max_det: # limit detections i = i[:max_det] if merge and (1 < n < 3E3): # Merge NMS (boxes merged using weighted mean) # update boxes as boxes(i,4) = weights(i,n) * boxes(n,4) iou = box_iou(boxes[i], boxes) > iou_thres # iou matrix weights = iou * scores[None] # box weights x[i, :4] = torch.mm(weights, x[:, :4]).float() / weights.sum(1, keepdim=True) # merged boxes if redundant: i = i[iou.sum(1) > 1] # require redundancy output[xi] = x[i] if (time.time() - t) > time_limit: LOGGER.warning(f'WARNING: NMS time limit {time_limit:.3f}s exceeded') break # time limit exceeded return output def strip_optimizer(f='best.pt', s=''): # from utils.general import *; strip_optimizer() # Strip optimizer from 'f' to finalize training, optionally save as 's' x = torch.load(f, map_location=torch.device('cpu')) if x.get('ema'): x['model'] = x['ema'] # replace model with ema for k in 'optimizer', 'best_fitness', 'wandb_id', 'ema', 'updates': # keys x[k] = None x['epoch'] = -1 x['model'].half() # to FP16 for p in x['model'].parameters(): p.requires_grad = False torch.save(x, s or f) mb = os.path.getsize(s or f) / 1E6 # filesize LOGGER.info(f"Optimizer stripped from {f},{(' saved as %s,' % s) if s else ''} {mb:.1f}MB") def print_mutation(results, hyp, save_dir, bucket, prefix=colorstr('evolve: ')): evolve_csv = save_dir / 'evolve.csv' evolve_yaml = save_dir / 'hyp_evolve.yaml' keys = ('metrics/precision', 'metrics/recall', 'metrics/mAP_0.5', 'metrics/mAP_0.5:0.95', 'val/box_loss', 'val/obj_loss', 'val/cls_loss') + tuple(hyp.keys()) # [results + hyps] keys = tuple(x.strip() for x in keys) vals = results + tuple(hyp.values()) n = len(keys) # Download (optional) if bucket: url = f'gs://{bucket}/evolve.csv' if gsutil_getsize(url) > (evolve_csv.stat().st_size if evolve_csv.exists() else 0): os.system(f'gsutil cp {url} {save_dir}') # download evolve.csv if larger than local # Log to evolve.csv s = '' if evolve_csv.exists() else (('%20s,' * n % keys).rstrip(',') + '\n') # add header with open(evolve_csv, 'a') as f: f.write(s + ('%20.5g,' * n % vals).rstrip(',') + '\n') # Save yaml with open(evolve_yaml, 'w') as f: data = pd.read_csv(evolve_csv) data = data.rename(columns=lambda x: x.strip()) # strip keys i = np.argmax(fitness(data.values[:, :4])) # generations = len(data) f.write('# YOLOv5 Hyperparameter Evolution Results\n' + f'# Best generation: {i}\n' + f'# Last generation: {generations - 1}\n' + '# ' + ', '.join(f'{x.strip():>20s}' for x in keys[:7]) + '\n' + '# ' + ', '.join(f'{x:>20.5g}' for x in data.values[i, :7]) + '\n\n') yaml.safe_dump(data.loc[i][7:].to_dict(), f, sort_keys=False) # Print to screen LOGGER.info(prefix + f'{generations} generations finished, current result:\n' + prefix + ', '.join(f'{x.strip():>20s}' for x in keys) + '\n' + prefix + ', '.join(f'{x:20.5g}' for x in vals) + '\n\n') if bucket: os.system(f'gsutil cp {evolve_csv} {evolve_yaml} gs://{bucket}') # upload def apply_classifier(x, model, img, im0): # Apply a second stage classifier to YOLO outputs # Example model = torchvision.models.__dict__['efficientnet_b0'](pretrained=True).to(device).eval() im0 = [im0] if isinstance(im0, np.ndarray) else im0 for i, d in enumerate(x): # per image if d is not None and len(d): d = d.clone() # Reshape and pad cutouts b = xyxy2xywh(d[:, :4]) # boxes b[:, 2:] = b[:, 2:].max(1)[0].unsqueeze(1) # rectangle to square b[:, 2:] = b[:, 2:] * 1.3 + 30 # pad d[:, :4] = xywh2xyxy(b).long() # Rescale boxes from img_size to im0 size scale_coords(img.shape[2:], d[:, :4], im0[i].shape) # Classes pred_cls1 = d[:, 5].long() ims = [] for j, a in enumerate(d): # per item cutout = im0[i][int(a[1]):int(a[3]), int(a[0]):int(a[2])] im = cv2.resize(cutout, (224, 224)) # BGR # cv2.imwrite('example%i.jpg' % j, cutout) im = im[:, :, ::-1].transpose(2, 0, 1) # BGR to RGB, to 3x416x416 im = np.ascontiguousarray(im, dtype=np.float32) # uint8 to float32 im /= 255 # 0 - 255 to 0.0 - 1.0 ims.append(im) pred_cls2 = model(torch.Tensor(ims).to(d.device)).argmax(1) # classifier prediction x[i] = x[i][pred_cls1 == pred_cls2] # retain matching class detections return x def increment_path(path, exist_ok=False, sep='', mkdir=False): # Increment file or directory path, i.e. runs/exp --> runs/exp{sep}2, runs/exp{sep}3, ... etc. path = Path(path) # os-agnostic if path.exists() and not exist_ok: path, suffix = (path.with_suffix(''), path.suffix) if path.is_file() else (path, '') dirs = glob.glob(f"{path}{sep}*") # similar paths matches = [re.search(rf"%s{sep}(\d+)" % path.stem, d) for d in dirs] i = [int(m.groups()[0]) for m in matches if m] # indices n = max(i) + 1 if i else 2 # increment number path = Path(f"{path}{sep}{n}{suffix}") # increment path if mkdir: path.mkdir(parents=True, exist_ok=True) # make directory return path # OpenCV Chinese-friendly functions ------------------------------------------------------------------------------------ imshow_ = cv2.imshow # copy to avoid recursion errors def imread(path, flags=cv2.IMREAD_COLOR): return cv2.imdecode(np.fromfile(path, np.uint8), flags) def imwrite(path, im): try: cv2.imencode(Path(path).suffix, im)[1].tofile(path) return True except Exception: return False def imshow(path, im): imshow_(path.encode('unicode_escape').decode(), im) cv2.imread, cv2.imwrite, cv2.imshow = imread, imwrite, imshow # redefine # Variables ------------------------------------------------------------------------------------------------------------ NCOLS = 0 if is_docker() else shutil.get_terminal_size().columns # terminal window size for tqdm ================================================ FILE: RStask/ObjectDetection/utils/google_app_engine/Dockerfile ================================================ FROM gcr.io/google-appengine/python # Create a virtualenv for dependencies. This isolates these packages from # system-level packages. # Use -p python3 or -p python3.7 to select python version. Default is version 2. RUN virtualenv /env -p python3 # Setting these environment variables are the same as running # source /env/bin/activate. ENV VIRTUAL_ENV /env ENV PATH /env/bin:$PATH RUN apt-get update && apt-get install -y python-opencv # Copy the application's requirements.txt and run pip to install all # dependencies into the virtualenv. ADD requirements.txt /app/requirements.txt RUN pip install -r /app/requirements.txt # Add the application source code. ADD . /app # Run a WSGI server to serve the application. gunicorn must be declared as # a dependency in requirements.txt. CMD gunicorn -b :$PORT main:app ================================================ FILE: RStask/ObjectDetection/utils/google_app_engine/additional_requirements.txt ================================================ # add these requirements in your app on top of the existing ones pip==21.1 Flask==1.0.2 gunicorn==19.9.0 ================================================ FILE: RStask/ObjectDetection/utils/google_app_engine/app.yaml ================================================ runtime: custom env: flex service: yolov5app liveness_check: initial_delay_sec: 600 manual_scaling: instances: 1 resources: cpu: 1 memory_gb: 4 disk_size_gb: 20 ================================================ FILE: RStask/ObjectDetection/utils/loggers/__init__.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Logging utils """ import os import warnings from threading import Thread import pkg_resources as pkg import torch from torch.utils.tensorboard import SummaryWriter from utils.general import colorstr, cv2, emojis from utils.loggers.wandb.wandb_utils import WandbLogger from utils.plots import plot_images, plot_results from utils.torch_utils import de_parallel LOGGERS = ('csv', 'tb', 'wandb') # text-file, TensorBoard, Weights & Biases RANK = int(os.getenv('RANK', -1)) try: import wandb assert hasattr(wandb, '__version__') # verify package import not local dir if pkg.parse_version(wandb.__version__) >= pkg.parse_version('0.12.2') and RANK in [0, -1]: try: wandb_login_success = wandb.login(timeout=30) except wandb.errors.UsageError: # known non-TTY terminal issue wandb_login_success = False if not wandb_login_success: wandb = None except (ImportError, AssertionError): wandb = None class Loggers(): # YOLOv5 Loggers class def __init__(self, save_dir=None, weights=None, opt=None, hyp=None, logger=None, include=LOGGERS): self.save_dir = save_dir self.weights = weights self.opt = opt self.hyp = hyp self.logger = logger # for printing results to console self.include = include self.keys = [ 'train/box_loss', 'train/obj_loss', 'train/cls_loss', # train loss 'metrics/precision', 'metrics/recall', 'metrics/mAP_0.5', 'metrics/mAP_0.5:0.95', # metrics 'val/box_loss', 'val/obj_loss', 'val/cls_loss', # val loss 'x/lr0', 'x/lr1', 'x/lr2'] # params self.best_keys = ['best/epoch', 'best/precision', 'best/recall', 'best/mAP_0.5', 'best/mAP_0.5:0.95'] for k in LOGGERS: setattr(self, k, None) # init empty logger dictionary self.csv = True # always log to csv # Message if not wandb: prefix = colorstr('Weights & Biases: ') s = f"{prefix}run 'pip install wandb' to automatically track and visualize YOLOv5 🚀 runs (RECOMMENDED)" self.logger.info(emojis(s)) # TensorBoard s = self.save_dir if 'tb' in self.include and not self.opt.evolve: prefix = colorstr('TensorBoard: ') self.logger.info(f"{prefix}Start with 'tensorboard --logdir {s.parent}', view at http://localhost:6006/") self.tb = SummaryWriter(str(s)) # W&B if wandb and 'wandb' in self.include: wandb_artifact_resume = isinstance(self.opt.resume, str) and self.opt.resume.startswith('wandb-artifact://') run_id = torch.load(self.weights).get('wandb_id') if self.opt.resume and not wandb_artifact_resume else None self.opt.hyp = self.hyp # add hyperparameters self.wandb = WandbLogger(self.opt, run_id) # temp warn. because nested artifacts not supported after 0.12.10 if pkg.parse_version(wandb.__version__) >= pkg.parse_version('0.12.11'): self.logger.warning( "YOLOv5 temporarily requires wandb version 0.12.10 or below. Some features may not work as expected." ) else: self.wandb = None def on_train_start(self): # Callback runs on train start pass def on_pretrain_routine_end(self): # Callback runs on pre-train routine end paths = self.save_dir.glob('*labels*.jpg') # training labels if self.wandb: self.wandb.log({"Labels": [wandb.Image(str(x), caption=x.name) for x in paths]}) def on_train_batch_end(self, ni, model, imgs, targets, paths, plots): # Callback runs on train batch end if plots: if ni == 0: if not self.opt.sync_bn: # --sync known issue https://github.com/ultralytics/yolov5/issues/3754 with warnings.catch_warnings(): warnings.simplefilter('ignore') # suppress jit trace warning self.tb.add_graph(torch.jit.trace(de_parallel(model), imgs[0:1], strict=False), []) if ni < 3: f = self.save_dir / f'train_batch{ni}.jpg' # filename Thread(target=plot_images, args=(imgs, targets, paths, f), daemon=True).start() if self.wandb and ni == 10: files = sorted(self.save_dir.glob('train*.jpg')) self.wandb.log({'Mosaics': [wandb.Image(str(f), caption=f.name) for f in files if f.exists()]}) def on_train_epoch_end(self, epoch): # Callback runs on train epoch end if self.wandb: self.wandb.current_epoch = epoch + 1 def on_val_image_end(self, pred, predn, path, names, im): # Callback runs on val image end if self.wandb: self.wandb.val_one_image(pred, predn, path, names, im) def on_val_end(self): # Callback runs on val end if self.wandb: files = sorted(self.save_dir.glob('val*.jpg')) self.wandb.log({"Validation": [wandb.Image(str(f), caption=f.name) for f in files]}) def on_fit_epoch_end(self, vals, epoch, best_fitness, fi): # Callback runs at the end of each fit (train+val) epoch x = {k: v for k, v in zip(self.keys, vals)} # dict if self.csv: file = self.save_dir / 'results.csv' n = len(x) + 1 # number of cols s = '' if file.exists() else (('%20s,' * n % tuple(['epoch'] + self.keys)).rstrip(',') + '\n') # add header with open(file, 'a') as f: f.write(s + ('%20.5g,' * n % tuple([epoch] + vals)).rstrip(',') + '\n') if self.tb: for k, v in x.items(): self.tb.add_scalar(k, v, epoch) if self.wandb: if best_fitness == fi: best_results = [epoch] + vals[3:7] for i, name in enumerate(self.best_keys): self.wandb.wandb_run.summary[name] = best_results[i] # log best results in the summary self.wandb.log(x) self.wandb.end_epoch(best_result=best_fitness == fi) def on_model_save(self, last, epoch, final_epoch, best_fitness, fi): # Callback runs on model save event if self.wandb: if ((epoch + 1) % self.opt.save_period == 0 and not final_epoch) and self.opt.save_period != -1: self.wandb.log_model(last.parent, self.opt, epoch, fi, best_model=best_fitness == fi) def on_train_end(self, last, best, plots, epoch, results): # Callback runs on training end if plots: plot_results(file=self.save_dir / 'results.csv') # save results.png files = ['results.png', 'confusion_matrix.png', *(f'{x}_curve.png' for x in ('F1', 'PR', 'P', 'R'))] files = [(self.save_dir / f) for f in files if (self.save_dir / f).exists()] # filter if self.tb: for f in files: self.tb.add_image(f.stem, cv2.imread(str(f))[..., ::-1], epoch, dataformats='HWC') if self.wandb: self.wandb.log({k: v for k, v in zip(self.keys[3:10], results)}) # log best.pt val results self.wandb.log({"Results": [wandb.Image(str(f), caption=f.name) for f in files]}) # Calling wandb.log. TODO: Refactor this into WandbLogger.log_model if not self.opt.evolve: wandb.log_artifact(str(best if best.exists() else last), type='model', name='run_' + self.wandb.wandb_run.id + '_model', aliases=['latest', 'best', 'stripped']) self.wandb.finish_run() def on_params_update(self, params): # Update hyperparams or configs of the experiment # params: A dict containing {param: value} pairs if self.wandb: self.wandb.wandb_run.config.update(params, allow_val_change=True) ================================================ FILE: RStask/ObjectDetection/utils/loggers/wandb/README.md ================================================ 📚 This guide explains how to use **Weights & Biases** (W&B) with YOLOv5 🚀. UPDATED 29 September 2021. - [About Weights & Biases](#about-weights-&-biases) - [First-Time Setup](#first-time-setup) - [Viewing runs](#viewing-runs) - [Disabling wandb](#disabling-wandb) - [Advanced Usage: Dataset Versioning and Evaluation](#advanced-usage) - [Reports: Share your work with the world!](#reports) ## About Weights & Biases Think of [W&B](https://wandb.ai/site?utm_campaign=repo_yolo_wandbtutorial) like GitHub for machine learning models. With a few lines of code, save everything you need to debug, compare and reproduce your models — architecture, hyperparameters, git commits, model weights, GPU usage, and even datasets and predictions. Used by top researchers including teams at OpenAI, Lyft, Github, and MILA, W&B is part of the new standard of best practices for machine learning. How W&B can help you optimize your machine learning workflows: - [Debug](https://wandb.ai/wandb/getting-started/reports/Visualize-Debug-Machine-Learning-Models--VmlldzoyNzY5MDk#Free-2) model performance in real time - [GPU usage](https://wandb.ai/wandb/getting-started/reports/Visualize-Debug-Machine-Learning-Models--VmlldzoyNzY5MDk#System-4) visualized automatically - [Custom charts](https://wandb.ai/wandb/customizable-charts/reports/Powerful-Custom-Charts-To-Debug-Model-Peformance--VmlldzoyNzY4ODI) for powerful, extensible visualization - [Share insights](https://wandb.ai/wandb/getting-started/reports/Visualize-Debug-Machine-Learning-Models--VmlldzoyNzY5MDk#Share-8) interactively with collaborators - [Optimize hyperparameters](https://docs.wandb.com/sweeps) efficiently - [Track](https://docs.wandb.com/artifacts) datasets, pipelines, and production models ## First-Time Setup
Toggle Details When you first train, W&B will prompt you to create a new account and will generate an **API key** for you. If you are an existing user you can retrieve your key from https://wandb.ai/authorize. This key is used to tell W&B where to log your data. You only need to supply your key once, and then it is remembered on the same device. W&B will create a cloud **project** (default is 'YOLOv5') for your training runs, and each new training run will be provided a unique run **name** within that project as project/name. You can also manually set your project and run name as: ```shell $ python train.py --project ... --name ... ``` YOLOv5 notebook example: Open In Colab Open In Kaggle Screen Shot 2021-09-29 at 10 23 13 PM
## Viewing Runs
Toggle Details Run information streams from your environment to the W&B cloud console as you train. This allows you to monitor and even cancel runs in realtime . All important information is logged: - Training & Validation losses - Metrics: Precision, Recall, mAP@0.5, mAP@0.5:0.95 - Learning Rate over time - A bounding box debugging panel, showing the training progress over time - GPU: Type, **GPU Utilization**, power, temperature, **CUDA memory usage** - System: Disk I/0, CPU utilization, RAM memory usage - Your trained model as W&B Artifact - Environment: OS and Python types, Git repository and state, **training command**

Weights & Biases dashboard

## Disabling wandb - training after running `wandb disabled` inside that directory creates no wandb run ![Screenshot (84)](https://user-images.githubusercontent.com/15766192/143441777-c780bdd7-7cb4-4404-9559-b4316030a985.png) - To enable wandb again, run `wandb online` ![Screenshot (85)](https://user-images.githubusercontent.com/15766192/143441866-7191b2cb-22f0-4e0f-ae64-2dc47dc13078.png) ## Advanced Usage You can leverage W&B artifacts and Tables integration to easily visualize and manage your datasets, models and training evaluations. Here are some quick examples to get you started.

1: Train and Log Evaluation simultaneousy

This is an extension of the previous section, but it'll also training after uploading the dataset. This also evaluation Table Evaluation table compares your predictions and ground truths across the validation set for each epoch. It uses the references to the already uploaded datasets, so no images will be uploaded from your system more than once.
Usage Code $ python train.py --upload_data val ![Screenshot from 2021-11-21 17-40-06](https://user-images.githubusercontent.com/15766192/142761183-c1696d8c-3f38-45ab-991a-bb0dfd98ae7d.png)

2. Visualize and Version Datasets

Log, visualize, dynamically query, and understand your data with W&B Tables. You can use the following command to log your dataset as a W&B Table. This will generate a {dataset}_wandb.yaml file which can be used to train from dataset artifact.
Usage Code $ python utils/logger/wandb/log_dataset.py --project ... --name ... --data .. ![Screenshot (64)](https://user-images.githubusercontent.com/15766192/128486078-d8433890-98a3-4d12-8986-b6c0e3fc64b9.png)

3: Train using dataset artifact

When you upload a dataset as described in the first section, you get a new config file with an added `_wandb` to its name. This file contains the information that can be used to train a model directly from the dataset artifact. This also logs evaluation
Usage Code $ python train.py --data {data}_wandb.yaml ![Screenshot (72)](https://user-images.githubusercontent.com/15766192/128979739-4cf63aeb-a76f-483f-8861-1c0100b938a5.png)

4: Save model checkpoints as artifacts

To enable saving and versioning checkpoints of your experiment, pass `--save_period n` with the base cammand, where `n` represents checkpoint interval. You can also log both the dataset and model checkpoints simultaneously. If not passed, only the final model will be logged
Usage Code $ python train.py --save_period 1 ![Screenshot (68)](https://user-images.githubusercontent.com/15766192/128726138-ec6c1f60-639d-437d-b4ee-3acd9de47ef3.png)

5: Resume runs from checkpoint artifacts.

Any run can be resumed using artifacts if the --resume argument starts with wandb-artifact:// prefix followed by the run path, i.e, wandb-artifact://username/project/runid . This doesn't require the model checkpoint to be present on the local system.
Usage Code $ python train.py --resume wandb-artifact://{run_path} ![Screenshot (70)](https://user-images.githubusercontent.com/15766192/128728988-4e84b355-6c87-41ae-a591-14aecf45343e.png)

6: Resume runs from dataset artifact & checkpoint artifacts.

Local dataset or model checkpoints are not required. This can be used to resume runs directly on a different device The syntax is same as the previous section, but you'll need to lof both the dataset and model checkpoints as artifacts, i.e, set bot --upload_dataset or train from _wandb.yaml file and set --save_period
Usage Code $ python train.py --resume wandb-artifact://{run_path} ![Screenshot (70)](https://user-images.githubusercontent.com/15766192/128728988-4e84b355-6c87-41ae-a591-14aecf45343e.png)

Reports

W&B Reports can be created from your saved runs for sharing online. Once a report is created you will receive a link you can use to publically share your results. Here is an example report created from the COCO128 tutorial trainings of all four YOLOv5 models ([link](https://wandb.ai/glenn-jocher/yolov5_tutorial/reports/YOLOv5-COCO128-Tutorial-Results--VmlldzozMDI5OTY)). Weights & Biases Reports ## Environments YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including [CUDA](https://developer.nvidia.com/cuda)/[CUDNN](https://developer.nvidia.com/cudnn), [Python](https://www.python.org/) and [PyTorch](https://pytorch.org/) preinstalled): - **Google Colab and Kaggle** notebooks with free GPU: Open In Colab Open In Kaggle - **Google Cloud** Deep Learning VM. See [GCP Quickstart Guide](https://github.com/ultralytics/yolov5/wiki/GCP-Quickstart) - **Amazon** Deep Learning AMI. See [AWS Quickstart Guide](https://github.com/ultralytics/yolov5/wiki/AWS-Quickstart) - **Docker Image**. See [Docker Quickstart Guide](https://github.com/ultralytics/yolov5/wiki/Docker-Quickstart) Docker Pulls ## Status ![CI CPU testing](https://github.com/ultralytics/yolov5/workflows/CI%20CPU%20testing/badge.svg) If this badge is green, all [YOLOv5 GitHub Actions](https://github.com/ultralytics/yolov5/actions) Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training ([train.py](https://github.com/ultralytics/yolov5/blob/master/train.py)), validation ([val.py](https://github.com/ultralytics/yolov5/blob/master/val.py)), inference ([detect.py](https://github.com/ultralytics/yolov5/blob/master/detect.py)) and export ([export.py](https://github.com/ultralytics/yolov5/blob/master/export.py)) on macOS, Windows, and Ubuntu every 24 hours and on every commit. ================================================ FILE: RStask/ObjectDetection/utils/loggers/wandb/__init__.py ================================================ ================================================ FILE: RStask/ObjectDetection/utils/loggers/wandb/log_dataset.py ================================================ import argparse from wandb_utils import WandbLogger from utils.general import LOGGER WANDB_ARTIFACT_PREFIX = 'wandb-artifact://' def create_dataset_artifact(opt): logger = WandbLogger(opt, None, job_type='Dataset Creation') # TODO: return value unused if not logger.wandb: LOGGER.info("install wandb using `pip install wandb` to log the dataset") if __name__ == '__main__': parser = argparse.ArgumentParser() parser.add_argument('--data', type=str, default='data/coco128.yaml', help='data.yaml path') parser.add_argument('--single-cls', action='store_true', help='train as single-class dataset') parser.add_argument('--project', type=str, default='YOLOv5', help='name of W&B Project') parser.add_argument('--entity', default=None, help='W&B entity') parser.add_argument('--name', type=str, default='log dataset', help='name of W&B run') opt = parser.parse_args() opt.resume = False # Explicitly disallow resume check for dataset upload job create_dataset_artifact(opt) ================================================ FILE: RStask/ObjectDetection/utils/loggers/wandb/sweep.py ================================================ import sys from pathlib import Path import wandb FILE = Path(__file__).resolve() ROOT = FILE.parents[3] # YOLOv5 root directory if str(ROOT) not in sys.path: sys.path.append(str(ROOT)) # add ROOT to PATH from train import parse_opt, train from utils.callbacks import Callbacks from utils.general import increment_path from utils.torch_utils import select_device def sweep(): wandb.init() # Get hyp dict from sweep agent. Copy because train() modifies parameters which confused wandb. hyp_dict = vars(wandb.config).get("_items").copy() # Workaround: get necessary opt args opt = parse_opt(known=True) opt.batch_size = hyp_dict.get("batch_size") opt.save_dir = str(increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok or opt.evolve)) opt.epochs = hyp_dict.get("epochs") opt.nosave = True opt.data = hyp_dict.get("data") opt.weights = str(opt.weights) opt.cfg = str(opt.cfg) opt.data = str(opt.data) opt.hyp = str(opt.hyp) opt.project = str(opt.project) device = select_device(opt.device, batch_size=opt.batch_size) # train train(hyp_dict, opt, device, callbacks=Callbacks()) if __name__ == "__main__": sweep() ================================================ FILE: RStask/ObjectDetection/utils/loggers/wandb/sweep.yaml ================================================ # Hyperparameters for training # To set range- # Provide min and max values as: # parameter: # # min: scalar # max: scalar # OR # # Set a specific list of search space- # parameter: # values: [scalar1, scalar2, scalar3...] # # You can use grid, bayesian and hyperopt search strategy # For more info on configuring sweeps visit - https://docs.wandb.ai/guides/sweeps/configuration program: utils/loggers/wandb/sweep.py method: random metric: name: metrics/mAP_0.5 goal: maximize parameters: # hyperparameters: set either min, max range or values list data: value: "data/coco128.yaml" batch_size: values: [64] epochs: values: [10] lr0: distribution: uniform min: 1e-5 max: 1e-1 lrf: distribution: uniform min: 0.01 max: 1.0 momentum: distribution: uniform min: 0.6 max: 0.98 weight_decay: distribution: uniform min: 0.0 max: 0.001 warmup_epochs: distribution: uniform min: 0.0 max: 5.0 warmup_momentum: distribution: uniform min: 0.0 max: 0.95 warmup_bias_lr: distribution: uniform min: 0.0 max: 0.2 box: distribution: uniform min: 0.02 max: 0.2 cls: distribution: uniform min: 0.2 max: 4.0 cls_pw: distribution: uniform min: 0.5 max: 2.0 obj: distribution: uniform min: 0.2 max: 4.0 obj_pw: distribution: uniform min: 0.5 max: 2.0 iou_t: distribution: uniform min: 0.1 max: 0.7 anchor_t: distribution: uniform min: 2.0 max: 8.0 fl_gamma: distribution: uniform min: 0.0 max: 4.0 hsv_h: distribution: uniform min: 0.0 max: 0.1 hsv_s: distribution: uniform min: 0.0 max: 0.9 hsv_v: distribution: uniform min: 0.0 max: 0.9 degrees: distribution: uniform min: 0.0 max: 45.0 translate: distribution: uniform min: 0.0 max: 0.9 scale: distribution: uniform min: 0.0 max: 0.9 shear: distribution: uniform min: 0.0 max: 10.0 perspective: distribution: uniform min: 0.0 max: 0.001 flipud: distribution: uniform min: 0.0 max: 1.0 fliplr: distribution: uniform min: 0.0 max: 1.0 mosaic: distribution: uniform min: 0.0 max: 1.0 mixup: distribution: uniform min: 0.0 max: 1.0 copy_paste: distribution: uniform min: 0.0 max: 1.0 ================================================ FILE: RStask/ObjectDetection/utils/loggers/wandb/wandb_utils.py ================================================ """Utilities and tools for tracking runs with Weights & Biases.""" import logging import os import sys from contextlib import contextmanager from pathlib import Path from typing import Dict import yaml from tqdm.auto import tqdm FILE = Path(__file__).resolve() ROOT = FILE.parents[3] # YOLOv5 root directory if str(ROOT) not in sys.path: sys.path.append(str(ROOT)) # add ROOT to PATH from utils.datasets import LoadImagesAndLabels, img2label_paths from utils.general import LOGGER, check_dataset, check_file try: import wandb assert hasattr(wandb, '__version__') # verify package import not local dir except (ImportError, AssertionError): wandb = None RANK = int(os.getenv('RANK', -1)) WANDB_ARTIFACT_PREFIX = 'wandb-artifact://' def remove_prefix(from_string, prefix=WANDB_ARTIFACT_PREFIX): return from_string[len(prefix):] def check_wandb_config_file(data_config_file): wandb_config = '_wandb.'.join(data_config_file.rsplit('.', 1)) # updated data.yaml path if Path(wandb_config).is_file(): return wandb_config return data_config_file def check_wandb_dataset(data_file): is_trainset_wandb_artifact = False is_valset_wandb_artifact = False if check_file(data_file) and data_file.endswith('.yaml'): with open(data_file, errors='ignore') as f: data_dict = yaml.safe_load(f) is_trainset_wandb_artifact = isinstance(data_dict['train'], str) and data_dict['train'].startswith(WANDB_ARTIFACT_PREFIX) is_valset_wandb_artifact = isinstance(data_dict['val'], str) and data_dict['val'].startswith(WANDB_ARTIFACT_PREFIX) if is_trainset_wandb_artifact or is_valset_wandb_artifact: return data_dict else: return check_dataset(data_file) def get_run_info(run_path): run_path = Path(remove_prefix(run_path, WANDB_ARTIFACT_PREFIX)) run_id = run_path.stem project = run_path.parent.stem entity = run_path.parent.parent.stem model_artifact_name = 'run_' + run_id + '_model' return entity, project, run_id, model_artifact_name def check_wandb_resume(opt): process_wandb_config_ddp_mode(opt) if RANK not in [-1, 0] else None if isinstance(opt.resume, str): if opt.resume.startswith(WANDB_ARTIFACT_PREFIX): if RANK not in [-1, 0]: # For resuming DDP runs entity, project, run_id, model_artifact_name = get_run_info(opt.resume) api = wandb.Api() artifact = api.artifact(entity + '/' + project + '/' + model_artifact_name + ':latest') modeldir = artifact.download() opt.weights = str(Path(modeldir) / "last.pt") return True return None def process_wandb_config_ddp_mode(opt): with open(check_file(opt.data), errors='ignore') as f: data_dict = yaml.safe_load(f) # data dict train_dir, val_dir = None, None if isinstance(data_dict['train'], str) and data_dict['train'].startswith(WANDB_ARTIFACT_PREFIX): api = wandb.Api() train_artifact = api.artifact(remove_prefix(data_dict['train']) + ':' + opt.artifact_alias) train_dir = train_artifact.download() train_path = Path(train_dir) / 'data/images/' data_dict['train'] = str(train_path) if isinstance(data_dict['val'], str) and data_dict['val'].startswith(WANDB_ARTIFACT_PREFIX): api = wandb.Api() val_artifact = api.artifact(remove_prefix(data_dict['val']) + ':' + opt.artifact_alias) val_dir = val_artifact.download() val_path = Path(val_dir) / 'data/images/' data_dict['val'] = str(val_path) if train_dir or val_dir: ddp_data_path = str(Path(val_dir) / 'wandb_local_data.yaml') with open(ddp_data_path, 'w') as f: yaml.safe_dump(data_dict, f) opt.data = ddp_data_path class WandbLogger(): """Log training runs, datasets, models, and predictions to Weights & Biases. This logger sends information to W&B at wandb.ai. By default, this information includes hyperparameters, system configuration and metrics, model metrics, and basic data metrics and analyses. By providing additional command line arguments to train.py, datasets, models and predictions can also be logged. For more on how this logger is used, see the Weights & Biases documentation: https://docs.wandb.com/guides/integrations/yolov5 """ def __init__(self, opt, run_id=None, job_type='Training'): """ - Initialize WandbLogger instance - Upload dataset if opt.upload_dataset is True - Setup trainig processes if job_type is 'Training' arguments: opt (namespace) -- Commandline arguments for this run run_id (str) -- Run ID of W&B run to be resumed job_type (str) -- To set the job_type for this run """ # Pre-training routine -- self.job_type = job_type self.wandb, self.wandb_run = wandb, None if not wandb else wandb.run self.val_artifact, self.train_artifact = None, None self.train_artifact_path, self.val_artifact_path = None, None self.result_artifact = None self.val_table, self.result_table = None, None self.bbox_media_panel_images = [] self.val_table_path_map = None self.max_imgs_to_log = 16 self.wandb_artifact_data_dict = None self.data_dict = None # It's more elegant to stick to 1 wandb.init call, # but useful config data is overwritten in the WandbLogger's wandb.init call if isinstance(opt.resume, str): # checks resume from artifact if opt.resume.startswith(WANDB_ARTIFACT_PREFIX): entity, project, run_id, model_artifact_name = get_run_info(opt.resume) model_artifact_name = WANDB_ARTIFACT_PREFIX + model_artifact_name assert wandb, 'install wandb to resume wandb runs' # Resume wandb-artifact:// runs here| workaround for not overwriting wandb.config self.wandb_run = wandb.init(id=run_id, project=project, entity=entity, resume='allow', allow_val_change=True) opt.resume = model_artifact_name elif self.wandb: self.wandb_run = wandb.init(config=opt, resume="allow", project='YOLOv5' if opt.project == 'runs/train' else Path(opt.project).stem, entity=opt.entity, name=opt.name if opt.name != 'exp' else None, job_type=job_type, id=run_id, allow_val_change=True) if not wandb.run else wandb.run if self.wandb_run: if self.job_type == 'Training': if opt.upload_dataset: if not opt.resume: self.wandb_artifact_data_dict = self.check_and_upload_dataset(opt) if opt.resume: # resume from artifact if isinstance(opt.resume, str) and opt.resume.startswith(WANDB_ARTIFACT_PREFIX): self.data_dict = dict(self.wandb_run.config.data_dict) else: # local resume self.data_dict = check_wandb_dataset(opt.data) else: self.data_dict = check_wandb_dataset(opt.data) self.wandb_artifact_data_dict = self.wandb_artifact_data_dict or self.data_dict # write data_dict to config. useful for resuming from artifacts. Do this only when not resuming. self.wandb_run.config.update({'data_dict': self.wandb_artifact_data_dict}, allow_val_change=True) self.setup_training(opt) if self.job_type == 'Dataset Creation': self.wandb_run.config.update({"upload_dataset": True}) self.data_dict = self.check_and_upload_dataset(opt) def check_and_upload_dataset(self, opt): """ Check if the dataset format is compatible and upload it as W&B artifact arguments: opt (namespace)-- Commandline arguments for current run returns: Updated dataset info dictionary where local dataset paths are replaced by WAND_ARFACT_PREFIX links. """ assert wandb, 'Install wandb to upload dataset' config_path = self.log_dataset_artifact(opt.data, opt.single_cls, 'YOLOv5' if opt.project == 'runs/train' else Path(opt.project).stem) with open(config_path, errors='ignore') as f: wandb_data_dict = yaml.safe_load(f) return wandb_data_dict def setup_training(self, opt): """ Setup the necessary processes for training YOLO models: - Attempt to download model checkpoint and dataset artifacts if opt.resume stats with WANDB_ARTIFACT_PREFIX - Update data_dict, to contain info of previous run if resumed and the paths of dataset artifact if downloaded - Setup log_dict, initialize bbox_interval arguments: opt (namespace) -- commandline arguments for this run """ self.log_dict, self.current_epoch = {}, 0 self.bbox_interval = opt.bbox_interval if isinstance(opt.resume, str): modeldir, _ = self.download_model_artifact(opt) if modeldir: self.weights = Path(modeldir) / "last.pt" config = self.wandb_run.config opt.weights, opt.save_period, opt.batch_size, opt.bbox_interval, opt.epochs, opt.hyp, opt.imgsz = str( self.weights), config.save_period, config.batch_size, config.bbox_interval, config.epochs,\ config.hyp, config.imgsz data_dict = self.data_dict if self.val_artifact is None: # If --upload_dataset is set, use the existing artifact, don't download self.train_artifact_path, self.train_artifact = self.download_dataset_artifact( data_dict.get('train'), opt.artifact_alias) self.val_artifact_path, self.val_artifact = self.download_dataset_artifact( data_dict.get('val'), opt.artifact_alias) if self.train_artifact_path is not None: train_path = Path(self.train_artifact_path) / 'data/images/' data_dict['train'] = str(train_path) if self.val_artifact_path is not None: val_path = Path(self.val_artifact_path) / 'data/images/' data_dict['val'] = str(val_path) if self.val_artifact is not None: self.result_artifact = wandb.Artifact("run_" + wandb.run.id + "_progress", "evaluation") columns = ["epoch", "id", "ground truth", "prediction"] columns.extend(self.data_dict['names']) self.result_table = wandb.Table(columns) self.val_table = self.val_artifact.get("val") if self.val_table_path_map is None: self.map_val_table_path() if opt.bbox_interval == -1: self.bbox_interval = opt.bbox_interval = (opt.epochs // 10) if opt.epochs > 10 else 1 if opt.evolve or opt.noplots: self.bbox_interval = opt.bbox_interval = opt.epochs + 1 # disable bbox_interval train_from_artifact = self.train_artifact_path is not None and self.val_artifact_path is not None # Update the the data_dict to point to local artifacts dir if train_from_artifact: self.data_dict = data_dict def download_dataset_artifact(self, path, alias): """ download the model checkpoint artifact if the path starts with WANDB_ARTIFACT_PREFIX arguments: path -- path of the dataset to be used for training alias (str)-- alias of the artifact to be download/used for training returns: (str, wandb.Artifact) -- path of the downladed dataset and it's corresponding artifact object if dataset is found otherwise returns (None, None) """ if isinstance(path, str) and path.startswith(WANDB_ARTIFACT_PREFIX): artifact_path = Path(remove_prefix(path, WANDB_ARTIFACT_PREFIX) + ":" + alias) dataset_artifact = wandb.use_artifact(artifact_path.as_posix().replace("\\", "/")) assert dataset_artifact is not None, "'Error: W&B dataset artifact doesn\'t exist'" datadir = dataset_artifact.download() return datadir, dataset_artifact return None, None def download_model_artifact(self, opt): """ download the model checkpoint artifact if the resume path starts with WANDB_ARTIFACT_PREFIX arguments: opt (namespace) -- Commandline arguments for this run """ if opt.resume.startswith(WANDB_ARTIFACT_PREFIX): model_artifact = wandb.use_artifact(remove_prefix(opt.resume, WANDB_ARTIFACT_PREFIX) + ":latest") assert model_artifact is not None, 'Error: W&B model artifact doesn\'t exist' modeldir = model_artifact.download() # epochs_trained = model_artifact.metadata.get('epochs_trained') total_epochs = model_artifact.metadata.get('total_epochs') is_finished = total_epochs is None assert not is_finished, 'training is finished, can only resume incomplete runs.' return modeldir, model_artifact return None, None def log_model(self, path, opt, epoch, fitness_score, best_model=False): """ Log the model checkpoint as W&B artifact arguments: path (Path) -- Path of directory containing the checkpoints opt (namespace) -- Command line arguments for this run epoch (int) -- Current epoch number fitness_score (float) -- fitness score for current epoch best_model (boolean) -- Boolean representing if the current checkpoint is the best yet. """ model_artifact = wandb.Artifact('run_' + wandb.run.id + '_model', type='model', metadata={ 'original_url': str(path), 'epochs_trained': epoch + 1, 'save period': opt.save_period, 'project': opt.project, 'total_epochs': opt.epochs, 'fitness_score': fitness_score}) model_artifact.add_file(str(path / 'last.pt'), name='last.pt') wandb.log_artifact(model_artifact, aliases=['latest', 'last', 'epoch ' + str(self.current_epoch), 'best' if best_model else '']) LOGGER.info(f"Saving model artifact on epoch {epoch + 1}") def log_dataset_artifact(self, data_file, single_cls, project, overwrite_config=False): """ Log the dataset as W&B artifact and return the new data file with W&B links arguments: data_file (str) -- the .yaml file with information about the dataset like - path, classes etc. single_class (boolean) -- train multi-class data as single-class project (str) -- project name. Used to construct the artifact path overwrite_config (boolean) -- overwrites the data.yaml file if set to true otherwise creates a new file with _wandb postfix. Eg -> data_wandb.yaml returns: the new .yaml file with artifact links. it can be used to start training directly from artifacts """ upload_dataset = self.wandb_run.config.upload_dataset log_val_only = isinstance(upload_dataset, str) and upload_dataset == 'val' self.data_dict = check_dataset(data_file) # parse and check data = dict(self.data_dict) nc, names = (1, ['item']) if single_cls else (int(data['nc']), data['names']) names = {k: v for k, v in enumerate(names)} # to index dictionary # log train set if not log_val_only: self.train_artifact = self.create_dataset_table(LoadImagesAndLabels(data['train'], rect=True, batch_size=1), names, name='train') if data.get('train') else None if data.get('train'): data['train'] = WANDB_ARTIFACT_PREFIX + str(Path(project) / 'train') self.val_artifact = self.create_dataset_table( LoadImagesAndLabels(data['val'], rect=True, batch_size=1), names, name='val') if data.get('val') else None if data.get('val'): data['val'] = WANDB_ARTIFACT_PREFIX + str(Path(project) / 'val') path = Path(data_file) # create a _wandb.yaml file with artifacts links if both train and test set are logged if not log_val_only: path = (path.stem if overwrite_config else path.stem + '_wandb') + '.yaml' # updated data.yaml path path = ROOT / 'data' / path data.pop('download', None) data.pop('path', None) with open(path, 'w') as f: yaml.safe_dump(data, f) LOGGER.info(f"Created dataset config file {path}") if self.job_type == 'Training': # builds correct artifact pipeline graph if not log_val_only: self.wandb_run.log_artifact( self.train_artifact) # calling use_artifact downloads the dataset. NOT NEEDED! self.wandb_run.use_artifact(self.val_artifact) self.val_artifact.wait() self.val_table = self.val_artifact.get('val') self.map_val_table_path() else: self.wandb_run.log_artifact(self.train_artifact) self.wandb_run.log_artifact(self.val_artifact) return path def map_val_table_path(self): """ Map the validation dataset Table like name of file -> it's id in the W&B Table. Useful for - referencing artifacts for evaluation. """ self.val_table_path_map = {} LOGGER.info("Mapping dataset") for i, data in enumerate(tqdm(self.val_table.data)): self.val_table_path_map[data[3]] = data[0] def create_dataset_table(self, dataset: LoadImagesAndLabels, class_to_id: Dict[int, str], name: str = 'dataset'): """ Create and return W&B artifact containing W&B Table of the dataset. arguments: dataset -- instance of LoadImagesAndLabels class used to iterate over the data to build Table class_to_id -- hash map that maps class ids to labels name -- name of the artifact returns: dataset artifact to be logged or used """ # TODO: Explore multiprocessing to slpit this loop parallely| This is essential for speeding up the the logging artifact = wandb.Artifact(name=name, type="dataset") img_files = tqdm([dataset.path]) if isinstance(dataset.path, str) and Path(dataset.path).is_dir() else None img_files = tqdm(dataset.im_files) if not img_files else img_files for img_file in img_files: if Path(img_file).is_dir(): artifact.add_dir(img_file, name='data/images') labels_path = 'labels'.join(dataset.path.rsplit('images', 1)) artifact.add_dir(labels_path, name='data/labels') else: artifact.add_file(img_file, name='data/images/' + Path(img_file).name) label_file = Path(img2label_paths([img_file])[0]) artifact.add_file(str(label_file), name='data/labels/' + label_file.name) if label_file.exists() else None table = wandb.Table(columns=["id", "train_image", "Classes", "name"]) class_set = wandb.Classes([{'id': id, 'name': name} for id, name in class_to_id.items()]) for si, (img, labels, paths, shapes) in enumerate(tqdm(dataset)): box_data, img_classes = [], {} for cls, *xywh in labels[:, 1:].tolist(): cls = int(cls) box_data.append({ "position": { "middle": [xywh[0], xywh[1]], "width": xywh[2], "height": xywh[3]}, "class_id": cls, "box_caption": "%s" % (class_to_id[cls])}) img_classes[cls] = class_to_id[cls] boxes = {"ground_truth": {"box_data": box_data, "class_labels": class_to_id}} # inference-space table.add_data(si, wandb.Image(paths, classes=class_set, boxes=boxes), list(img_classes.values()), Path(paths).name) artifact.add(table, name) return artifact def log_training_progress(self, predn, path, names): """ Build evaluation Table. Uses reference from validation dataset table. arguments: predn (list): list of predictions in the native space in the format - [xmin, ymin, xmax, ymax, confidence, class] path (str): local path of the current evaluation image names (dict(int, str)): hash map that maps class ids to labels """ class_set = wandb.Classes([{'id': id, 'name': name} for id, name in names.items()]) box_data = [] avg_conf_per_class = [0] * len(self.data_dict['names']) pred_class_count = {} for *xyxy, conf, cls in predn.tolist(): if conf >= 0.25: cls = int(cls) box_data.append({ "position": { "minX": xyxy[0], "minY": xyxy[1], "maxX": xyxy[2], "maxY": xyxy[3]}, "class_id": cls, "box_caption": f"{names[cls]} {conf:.3f}", "scores": { "class_score": conf}, "domain": "pixel"}) avg_conf_per_class[cls] += conf if cls in pred_class_count: pred_class_count[cls] += 1 else: pred_class_count[cls] = 1 for pred_class in pred_class_count.keys(): avg_conf_per_class[pred_class] = avg_conf_per_class[pred_class] / pred_class_count[pred_class] boxes = {"predictions": {"box_data": box_data, "class_labels": names}} # inference-space id = self.val_table_path_map[Path(path).name] self.result_table.add_data(self.current_epoch, id, self.val_table.data[id][1], wandb.Image(self.val_table.data[id][1], boxes=boxes, classes=class_set), *avg_conf_per_class) def val_one_image(self, pred, predn, path, names, im): """ Log validation data for one image. updates the result Table if validation dataset is uploaded and log bbox media panel arguments: pred (list): list of scaled predictions in the format - [xmin, ymin, xmax, ymax, confidence, class] predn (list): list of predictions in the native space - [xmin, ymin, xmax, ymax, confidence, class] path (str): local path of the current evaluation image """ if self.val_table and self.result_table: # Log Table if Val dataset is uploaded as artifact self.log_training_progress(predn, path, names) if len(self.bbox_media_panel_images) < self.max_imgs_to_log and self.current_epoch > 0: if self.current_epoch % self.bbox_interval == 0: box_data = [{ "position": { "minX": xyxy[0], "minY": xyxy[1], "maxX": xyxy[2], "maxY": xyxy[3]}, "class_id": int(cls), "box_caption": f"{names[int(cls)]} {conf:.3f}", "scores": { "class_score": conf}, "domain": "pixel"} for *xyxy, conf, cls in pred.tolist()] boxes = {"predictions": {"box_data": box_data, "class_labels": names}} # inference-space self.bbox_media_panel_images.append(wandb.Image(im, boxes=boxes, caption=path.name)) def log(self, log_dict): """ save the metrics to the logging dictionary arguments: log_dict (Dict) -- metrics/media to be logged in current step """ if self.wandb_run: for key, value in log_dict.items(): self.log_dict[key] = value def end_epoch(self, best_result=False): """ commit the log_dict, model artifacts and Tables to W&B and flush the log_dict. arguments: best_result (boolean): Boolean representing if the result of this evaluation is best or not """ if self.wandb_run: with all_logging_disabled(): if self.bbox_media_panel_images: self.log_dict["BoundingBoxDebugger"] = self.bbox_media_panel_images try: wandb.log(self.log_dict) except BaseException as e: LOGGER.info( f"An error occurred in wandb logger. The training will proceed without interruption. More info\n{e}" ) self.wandb_run.finish() self.wandb_run = None self.log_dict = {} self.bbox_media_panel_images = [] if self.result_artifact: self.result_artifact.add(self.result_table, 'result') wandb.log_artifact(self.result_artifact, aliases=[ 'latest', 'last', 'epoch ' + str(self.current_epoch), ('best' if best_result else '')]) wandb.log({"evaluation": self.result_table}) columns = ["epoch", "id", "ground truth", "prediction"] columns.extend(self.data_dict['names']) self.result_table = wandb.Table(columns) self.result_artifact = wandb.Artifact("run_" + wandb.run.id + "_progress", "evaluation") def finish_run(self): """ Log metrics if any and finish the current W&B run """ if self.wandb_run: if self.log_dict: with all_logging_disabled(): wandb.log(self.log_dict) wandb.run.finish() @contextmanager def all_logging_disabled(highest_level=logging.CRITICAL): """ source - https://gist.github.com/simon-weber/7853144 A context manager that will prevent any logging messages triggered during the body from being processed. :param highest_level: the maximum logging level in use. This would only need to be changed if a custom level greater than CRITICAL is defined. """ previous_level = logging.root.manager.disable logging.disable(highest_level) try: yield finally: logging.disable(previous_level) ================================================ FILE: RStask/ObjectDetection/utils/loss.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Loss functions """ import torch import torch.nn as nn from utils.metrics import bbox_iou from utils.torch_utils import de_parallel def smooth_BCE(eps=0.1): # https://github.com/ultralytics/yolov3/issues/238#issuecomment-598028441 # return positive, negative label smoothing BCE targets return 1.0 - 0.5 * eps, 0.5 * eps class BCEBlurWithLogitsLoss(nn.Module): # BCEwithLogitLoss() with reduced missing label effects. def __init__(self, alpha=0.05): super().__init__() self.loss_fcn = nn.BCEWithLogitsLoss(reduction='none') # must be nn.BCEWithLogitsLoss() self.alpha = alpha def forward(self, pred, true): loss = self.loss_fcn(pred, true) pred = torch.sigmoid(pred) # prob from logits dx = pred - true # reduce only missing label effects # dx = (pred - true).abs() # reduce missing label and false label effects alpha_factor = 1 - torch.exp((dx - 1) / (self.alpha + 1e-4)) loss *= alpha_factor return loss.mean() class FocalLoss(nn.Module): # Wraps focal loss around existing loss_fcn(), i.e. criteria = FocalLoss(nn.BCEWithLogitsLoss(), gamma=1.5) def __init__(self, loss_fcn, gamma=1.5, alpha=0.25): super().__init__() self.loss_fcn = loss_fcn # must be nn.BCEWithLogitsLoss() self.gamma = gamma self.alpha = alpha self.reduction = loss_fcn.reduction self.loss_fcn.reduction = 'none' # required to apply FL to each element def forward(self, pred, true): loss = self.loss_fcn(pred, true) # p_t = torch.exp(-loss) # loss *= self.alpha * (1.000001 - p_t) ** self.gamma # non-zero power for gradient stability # TF implementation https://github.com/tensorflow/addons/blob/v0.7.1/tensorflow_addons/losses/focal_loss.py pred_prob = torch.sigmoid(pred) # prob from logits p_t = true * pred_prob + (1 - true) * (1 - pred_prob) alpha_factor = true * self.alpha + (1 - true) * (1 - self.alpha) modulating_factor = (1.0 - p_t) ** self.gamma loss *= alpha_factor * modulating_factor if self.reduction == 'mean': return loss.mean() elif self.reduction == 'sum': return loss.sum() else: # 'none' return loss class QFocalLoss(nn.Module): # Wraps Quality focal loss around existing loss_fcn(), i.e. criteria = FocalLoss(nn.BCEWithLogitsLoss(), gamma=1.5) def __init__(self, loss_fcn, gamma=1.5, alpha=0.25): super().__init__() self.loss_fcn = loss_fcn # must be nn.BCEWithLogitsLoss() self.gamma = gamma self.alpha = alpha self.reduction = loss_fcn.reduction self.loss_fcn.reduction = 'none' # required to apply FL to each element def forward(self, pred, true): loss = self.loss_fcn(pred, true) pred_prob = torch.sigmoid(pred) # prob from logits alpha_factor = true * self.alpha + (1 - true) * (1 - self.alpha) modulating_factor = torch.abs(true - pred_prob) ** self.gamma loss *= alpha_factor * modulating_factor if self.reduction == 'mean': return loss.mean() elif self.reduction == 'sum': return loss.sum() else: # 'none' return loss class ComputeLoss: sort_obj_iou = False # Compute losses def __init__(self, model, autobalance=False): device = next(model.parameters()).device # get model device h = model.hyp # hyperparameters # Define criteria BCEcls = nn.BCEWithLogitsLoss(pos_weight=torch.tensor([h['cls_pw']], device=device)) BCEobj = nn.BCEWithLogitsLoss(pos_weight=torch.tensor([h['obj_pw']], device=device)) # Class label smoothing https://arxiv.org/pdf/1902.04103.pdf eqn 3 self.cp, self.cn = smooth_BCE(eps=h.get('label_smoothing', 0.0)) # positive, negative BCE targets # Focal loss g = h['fl_gamma'] # focal loss gamma if g > 0: BCEcls, BCEobj = FocalLoss(BCEcls, g), FocalLoss(BCEobj, g) m = de_parallel(model).model[-1] # Detect() module self.balance = {3: [4.0, 1.0, 0.4]}.get(m.nl, [4.0, 1.0, 0.25, 0.06, 0.02]) # P3-P7 self.ssi = list(m.stride).index(16) if autobalance else 0 # stride 16 index self.BCEcls, self.BCEobj, self.gr, self.hyp, self.autobalance = BCEcls, BCEobj, 1.0, h, autobalance self.na = m.na # number of anchors self.nc = m.nc # number of classes self.nl = m.nl # number of layers self.anchors = m.anchors self.device = device def __call__(self, p, targets): # predictions, targets lcls = torch.zeros(1, device=self.device) # class loss lbox = torch.zeros(1, device=self.device) # box loss lobj = torch.zeros(1, device=self.device) # object loss tcls, tbox, indices, anchors = self.build_targets(p, targets) # targets # Losses for i, pi in enumerate(p): # layer index, layer predictions b, a, gj, gi = indices[i] # image, anchor, gridy, gridx tobj = torch.zeros(pi.shape[:4], dtype=pi.dtype, device=self.device) # target obj n = b.shape[0] # number of targets if n: # pxy, pwh, _, pcls = pi[b, a, gj, gi].tensor_split((2, 4, 5), dim=1) # faster, requires torch 1.8.0 pxy, pwh, _, pcls = pi[b, a, gj, gi].split((2, 2, 1, self.nc), 1) # target-subset of predictions # Regression pxy = pxy.sigmoid() * 2 - 0.5 pwh = (pwh.sigmoid() * 2) ** 2 * anchors[i] pbox = torch.cat((pxy, pwh), 1) # predicted box iou = bbox_iou(pbox, tbox[i], CIoU=True).squeeze() # iou(prediction, target) lbox += (1.0 - iou).mean() # iou loss # Objectness iou = iou.detach().clamp(0).type(tobj.dtype) if self.sort_obj_iou: j = iou.argsort() b, a, gj, gi, iou = b[j], a[j], gj[j], gi[j], iou[j] if self.gr < 1: iou = (1.0 - self.gr) + self.gr * iou tobj[b, a, gj, gi] = iou # iou ratio # Classification if self.nc > 1: # cls loss (only if multiple classes) t = torch.full_like(pcls, self.cn, device=self.device) # targets t[range(n), tcls[i]] = self.cp lcls += self.BCEcls(pcls, t) # BCE # Append targets to text file # with open('targets.txt', 'a') as file: # [file.write('%11.5g ' * 4 % tuple(x) + '\n') for x in torch.cat((txy[i], twh[i]), 1)] obji = self.BCEobj(pi[..., 4], tobj) lobj += obji * self.balance[i] # obj loss if self.autobalance: self.balance[i] = self.balance[i] * 0.9999 + 0.0001 / obji.detach().item() if self.autobalance: self.balance = [x / self.balance[self.ssi] for x in self.balance] lbox *= self.hyp['box'] lobj *= self.hyp['obj'] lcls *= self.hyp['cls'] bs = tobj.shape[0] # batch size return (lbox + lobj + lcls) * bs, torch.cat((lbox, lobj, lcls)).detach() def build_targets(self, p, targets): # Build targets for compute_loss(), input targets(image,class,x,y,w,h) na, nt = self.na, targets.shape[0] # number of anchors, targets tcls, tbox, indices, anch = [], [], [], [] gain = torch.ones(7, device=self.device) # normalized to gridspace gain ai = torch.arange(na, device=self.device).float().view(na, 1).repeat(1, nt) # same as .repeat_interleave(nt) targets = torch.cat((targets.repeat(na, 1, 1), ai[..., None]), 2) # append anchor indices g = 0.5 # bias off = torch.tensor( [ [0, 0], [1, 0], [0, 1], [-1, 0], [0, -1], # j,k,l,m # [1, 1], [1, -1], [-1, 1], [-1, -1], # jk,jm,lk,lm ], device=self.device).float() * g # offsets for i in range(self.nl): anchors = self.anchors[i] gain[2:6] = torch.tensor(p[i].shape)[[3, 2, 3, 2]] # xyxy gain # Match targets to anchors t = targets * gain # shape(3,n,7) if nt: # Matches r = t[..., 4:6] / anchors[:, None] # wh ratio j = torch.max(r, 1 / r).max(2)[0] < self.hyp['anchor_t'] # compare # j = wh_iou(anchors, t[:, 4:6]) > model.hyp['iou_t'] # iou(3,n)=wh_iou(anchors(3,2), gwh(n,2)) t = t[j] # filter # Offsets gxy = t[:, 2:4] # grid xy gxi = gain[[2, 3]] - gxy # inverse j, k = ((gxy % 1 < g) & (gxy > 1)).T l, m = ((gxi % 1 < g) & (gxi > 1)).T j = torch.stack((torch.ones_like(j), j, k, l, m)) t = t.repeat((5, 1, 1))[j] offsets = (torch.zeros_like(gxy)[None] + off[:, None])[j] else: t = targets[0] offsets = 0 # Define bc, gxy, gwh, a = t.chunk(4, 1) # (image, class), grid xy, grid wh, anchors a, (b, c) = a.long().view(-1), bc.long().T # anchors, image, class gij = (gxy - offsets).long() gi, gj = gij.T # grid indices # Append indices.append((b, a, gj.clamp_(0, gain[3] - 1), gi.clamp_(0, gain[2] - 1))) # image, anchor, grid indices tbox.append(torch.cat((gxy - gij, gwh), 1)) # box anch.append(anchors[a]) # anchors tcls.append(c) # class return tcls, tbox, indices, anch ================================================ FILE: RStask/ObjectDetection/utils/metrics.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Model validation metrics """ import math import warnings from pathlib import Path import matplotlib.pyplot as plt import numpy as np import torch def fitness(x): # Model fitness as a weighted combination of metrics w = [0.0, 0.0, 0.1, 0.9] # weights for [P, R, mAP@0.5, mAP@0.5:0.95] return (x[:, :4] * w).sum(1) def ap_per_class(tp, conf, pred_cls, target_cls, plot=False, save_dir='.', names=(), eps=1e-16): """ Compute the average precision, given the recall and precision curves. Source: https://github.com/rafaelpadilla/Object-Detection-Metrics. # Arguments tp: True positives (nparray, nx1 or nx10). conf: Objectness value from 0-1 (nparray). pred_cls: Predicted object classes (nparray). target_cls: True object classes (nparray). plot: Plot precision-recall curve at mAP@0.5 save_dir: Plot save directory # Returns The average precision as computed in py-faster-rcnn. """ # Sort by objectness i = np.argsort(-conf) tp, conf, pred_cls = tp[i], conf[i], pred_cls[i] # Find unique classes unique_classes, nt = np.unique(target_cls, return_counts=True) nc = unique_classes.shape[0] # number of classes, number of detections # Create Precision-Recall curve and compute AP for each class px, py = np.linspace(0, 1, 1000), [] # for plotting ap, p, r = np.zeros((nc, tp.shape[1])), np.zeros((nc, 1000)), np.zeros((nc, 1000)) for ci, c in enumerate(unique_classes): i = pred_cls == c n_l = nt[ci] # number of labels n_p = i.sum() # number of predictions if n_p == 0 or n_l == 0: continue else: # Accumulate FPs and TPs fpc = (1 - tp[i]).cumsum(0) tpc = tp[i].cumsum(0) # Recall recall = tpc / (n_l + eps) # recall curve r[ci] = np.interp(-px, -conf[i], recall[:, 0], left=0) # negative x, xp because xp decreases # Precision precision = tpc / (tpc + fpc) # precision curve p[ci] = np.interp(-px, -conf[i], precision[:, 0], left=1) # p at pr_score # AP from recall-precision curve for j in range(tp.shape[1]): ap[ci, j], mpre, mrec = compute_ap(recall[:, j], precision[:, j]) if plot and j == 0: py.append(np.interp(px, mrec, mpre)) # precision at mAP@0.5 # Compute F1 (harmonic mean of precision and recall) f1 = 2 * p * r / (p + r + eps) names = [v for k, v in names.items() if k in unique_classes] # list: only classes that have data names = {i: v for i, v in enumerate(names)} # to dict if plot: plot_pr_curve(px, py, ap, Path(save_dir) / 'PR_curve.png', names) plot_mc_curve(px, f1, Path(save_dir) / 'F1_curve.png', names, ylabel='F1') plot_mc_curve(px, p, Path(save_dir) / 'P_curve.png', names, ylabel='Precision') plot_mc_curve(px, r, Path(save_dir) / 'R_curve.png', names, ylabel='Recall') i = f1.mean(0).argmax() # max F1 index p, r, f1 = p[:, i], r[:, i], f1[:, i] tp = (r * nt).round() # true positives fp = (tp / (p + eps) - tp).round() # false positives return tp, fp, p, r, f1, ap, unique_classes.astype('int32') def compute_ap(recall, precision): """ Compute the average precision, given the recall and precision curves # Arguments recall: The recall curve (list) precision: The precision curve (list) # Returns Average precision, precision curve, recall curve """ # Append sentinel values to beginning and end mrec = np.concatenate(([0.0], recall, [1.0])) mpre = np.concatenate(([1.0], precision, [0.0])) # Compute the precision envelope mpre = np.flip(np.maximum.accumulate(np.flip(mpre))) # Integrate area under curve method = 'interp' # methods: 'continuous', 'interp' if method == 'interp': x = np.linspace(0, 1, 101) # 101-point interp (COCO) ap = np.trapz(np.interp(x, mrec, mpre), x) # integrate else: # 'continuous' i = np.where(mrec[1:] != mrec[:-1])[0] # points where x axis (recall) changes ap = np.sum((mrec[i + 1] - mrec[i]) * mpre[i + 1]) # area under curve return ap, mpre, mrec class ConfusionMatrix: # Updated version of https://github.com/kaanakan/object_detection_confusion_matrix def __init__(self, nc, conf=0.25, iou_thres=0.45): self.matrix = np.zeros((nc + 1, nc + 1)) self.nc = nc # number of classes self.conf = conf self.iou_thres = iou_thres def process_batch(self, detections, labels): """ Return intersection-over-union (Jaccard index) of boxes. Both sets of boxes are expected to be in (x1, y1, x2, y2) format. Arguments: detections (Array[N, 6]), x1, y1, x2, y2, conf, class labels (Array[M, 5]), class, x1, y1, x2, y2 Returns: None, updates confusion matrix accordingly """ detections = detections[detections[:, 4] > self.conf] gt_classes = labels[:, 0].int() detection_classes = detections[:, 5].int() iou = box_iou(labels[:, 1:], detections[:, :4]) x = torch.where(iou > self.iou_thres) if x[0].shape[0]: matches = torch.cat((torch.stack(x, 1), iou[x[0], x[1]][:, None]), 1).cpu().numpy() if x[0].shape[0] > 1: matches = matches[matches[:, 2].argsort()[::-1]] matches = matches[np.unique(matches[:, 1], return_index=True)[1]] matches = matches[matches[:, 2].argsort()[::-1]] matches = matches[np.unique(matches[:, 0], return_index=True)[1]] else: matches = np.zeros((0, 3)) n = matches.shape[0] > 0 m0, m1, _ = matches.transpose().astype(np.int16) for i, gc in enumerate(gt_classes): j = m0 == i if n and sum(j) == 1: self.matrix[detection_classes[m1[j]], gc] += 1 # correct else: self.matrix[self.nc, gc] += 1 # background FP if n: for i, dc in enumerate(detection_classes): if not any(m1 == i): self.matrix[dc, self.nc] += 1 # background FN def matrix(self): return self.matrix def tp_fp(self): tp = self.matrix.diagonal() # true positives fp = self.matrix.sum(1) - tp # false positives # fn = self.matrix.sum(0) - tp # false negatives (missed detections) return tp[:-1], fp[:-1] # remove background class def plot(self, normalize=True, save_dir='', names=()): try: import seaborn as sn array = self.matrix / ((self.matrix.sum(0).reshape(1, -1) + 1E-9) if normalize else 1) # normalize columns array[array < 0.005] = np.nan # don't annotate (would appear as 0.00) fig = plt.figure(figsize=(12, 9), tight_layout=True) nc, nn = self.nc, len(names) # number of classes, names sn.set(font_scale=1.0 if nc < 50 else 0.8) # for label size labels = (0 < nn < 99) and (nn == nc) # apply names to ticklabels with warnings.catch_warnings(): warnings.simplefilter('ignore') # suppress empty matrix RuntimeWarning: All-NaN slice encountered sn.heatmap(array, annot=nc < 30, annot_kws={ "size": 8}, cmap='Blues', fmt='.2f', square=True, vmin=0.0, xticklabels=names + ['background FP'] if labels else "auto", yticklabels=names + ['background FN'] if labels else "auto").set_facecolor((1, 1, 1)) fig.axes[0].set_xlabel('True') fig.axes[0].set_ylabel('Predicted') fig.savefig(Path(save_dir) / 'confusion_matrix.png', dpi=250) plt.close() except Exception as e: print(f'WARNING: ConfusionMatrix plot failure: {e}') def print(self): for i in range(self.nc + 1): print(' '.join(map(str, self.matrix[i]))) def bbox_iou(box1, box2, xywh=True, GIoU=False, DIoU=False, CIoU=False, eps=1e-7): # Returns Intersection over Union (IoU) of box1(1,4) to box2(n,4) # Get the coordinates of bounding boxes if xywh: # transform from xywh to xyxy (x1, y1, w1, h1), (x2, y2, w2, h2) = box1.chunk(4, 1), box2.chunk(4, 1) w1_, h1_, w2_, h2_ = w1 / 2, h1 / 2, w2 / 2, h2 / 2 b1_x1, b1_x2, b1_y1, b1_y2 = x1 - w1_, x1 + w1_, y1 - h1_, y1 + h1_ b2_x1, b2_x2, b2_y1, b2_y2 = x2 - w2_, x2 + w2_, y2 - h2_, y2 + h2_ else: # x1, y1, x2, y2 = box1 b1_x1, b1_y1, b1_x2, b1_y2 = box1.chunk(4, 1) b2_x1, b2_y1, b2_x2, b2_y2 = box2.chunk(4, 1) w1, h1 = b1_x2 - b1_x1, b1_y2 - b1_y1 + eps w2, h2 = b2_x2 - b2_x1, b2_y2 - b2_y1 + eps # Intersection area inter = (torch.min(b1_x2, b2_x2) - torch.max(b1_x1, b2_x1)).clamp(0) * \ (torch.min(b1_y2, b2_y2) - torch.max(b1_y1, b2_y1)).clamp(0) # Union Area union = w1 * h1 + w2 * h2 - inter + eps # IoU iou = inter / union if CIoU or DIoU or GIoU: cw = torch.max(b1_x2, b2_x2) - torch.min(b1_x1, b2_x1) # convex (smallest enclosing box) width ch = torch.max(b1_y2, b2_y2) - torch.min(b1_y1, b2_y1) # convex height if CIoU or DIoU: # Distance or Complete IoU https://arxiv.org/abs/1911.08287v1 c2 = cw ** 2 + ch ** 2 + eps # convex diagonal squared rho2 = ((b2_x1 + b2_x2 - b1_x1 - b1_x2) ** 2 + (b2_y1 + b2_y2 - b1_y1 - b1_y2) ** 2) / 4 # center dist ** 2 if CIoU: # https://github.com/Zzh-tju/DIoU-SSD-pytorch/blob/master/utils/box/box_utils.py#L47 v = (4 / math.pi ** 2) * torch.pow(torch.atan(w2 / h2) - torch.atan(w1 / h1), 2) with torch.no_grad(): alpha = v / (v - iou + (1 + eps)) return iou - (rho2 / c2 + v * alpha) # CIoU return iou - rho2 / c2 # DIoU c_area = cw * ch + eps # convex area return iou - (c_area - union) / c_area # GIoU https://arxiv.org/pdf/1902.09630.pdf return iou # IoU def box_area(box): # box = xyxy(4,n) return (box[2] - box[0]) * (box[3] - box[1]) def box_iou(box1, box2): # https://github.com/pytorch/vision/blob/master/torchvision/ops/boxes.py """ Return intersection-over-union (Jaccard index) of boxes. Both sets of boxes are expected to be in (x1, y1, x2, y2) format. Arguments: box1 (Tensor[N, 4]) box2 (Tensor[M, 4]) Returns: iou (Tensor[N, M]): the NxM matrix containing the pairwise IoU values for every element in boxes1 and boxes2 """ # inter(N,M) = (rb(N,M,2) - lt(N,M,2)).clamp(0).prod(2) (a1, a2), (b1, b2) = box1[:, None].chunk(2, 2), box2.chunk(2, 1) inter = (torch.min(a2, b2) - torch.max(a1, b1)).clamp(0).prod(2) # IoU = inter / (area1 + area2 - inter) return inter / (box_area(box1.T)[:, None] + box_area(box2.T) - inter) def bbox_ioa(box1, box2, eps=1E-7): """ Returns the intersection over box2 area given box1, box2. Boxes are x1y1x2y2 box1: np.array of shape(4) box2: np.array of shape(nx4) returns: np.array of shape(n) """ # Get the coordinates of bounding boxes b1_x1, b1_y1, b1_x2, b1_y2 = box1 b2_x1, b2_y1, b2_x2, b2_y2 = box2.T # Intersection area inter_area = (np.minimum(b1_x2, b2_x2) - np.maximum(b1_x1, b2_x1)).clip(0) * \ (np.minimum(b1_y2, b2_y2) - np.maximum(b1_y1, b2_y1)).clip(0) # box2 area box2_area = (b2_x2 - b2_x1) * (b2_y2 - b2_y1) + eps # Intersection over box2 area return inter_area / box2_area def wh_iou(wh1, wh2): # Returns the nxm IoU matrix. wh1 is nx2, wh2 is mx2 wh1 = wh1[:, None] # [N,1,2] wh2 = wh2[None] # [1,M,2] inter = torch.min(wh1, wh2).prod(2) # [N,M] return inter / (wh1.prod(2) + wh2.prod(2) - inter) # iou = inter / (area1 + area2 - inter) # Plots ---------------------------------------------------------------------------------------------------------------- def plot_pr_curve(px, py, ap, save_dir='pr_curve.png', names=()): # Precision-recall curve fig, ax = plt.subplots(1, 1, figsize=(9, 6), tight_layout=True) py = np.stack(py, axis=1) if 0 < len(names) < 21: # display per-class legend if < 21 classes for i, y in enumerate(py.T): ax.plot(px, y, linewidth=1, label=f'{names[i]} {ap[i, 0]:.3f}') # plot(recall, precision) else: ax.plot(px, py, linewidth=1, color='grey') # plot(recall, precision) ax.plot(px, py.mean(1), linewidth=3, color='blue', label='all classes %.3f mAP@0.5' % ap[:, 0].mean()) ax.set_xlabel('Recall') ax.set_ylabel('Precision') ax.set_xlim(0, 1) ax.set_ylim(0, 1) plt.legend(bbox_to_anchor=(1.04, 1), loc="upper left") fig.savefig(Path(save_dir), dpi=250) plt.close() def plot_mc_curve(px, py, save_dir='mc_curve.png', names=(), xlabel='Confidence', ylabel='Metric'): # Metric-confidence curve fig, ax = plt.subplots(1, 1, figsize=(9, 6), tight_layout=True) if 0 < len(names) < 21: # display per-class legend if < 21 classes for i, y in enumerate(py): ax.plot(px, y, linewidth=1, label=f'{names[i]}') # plot(confidence, metric) else: ax.plot(px, py.T, linewidth=1, color='grey') # plot(confidence, metric) y = py.mean(0) ax.plot(px, y, linewidth=3, color='blue', label=f'all classes {y.max():.2f} at {px[y.argmax()]:.3f}') ax.set_xlabel(xlabel) ax.set_ylabel(ylabel) ax.set_xlim(0, 1) ax.set_ylim(0, 1) plt.legend(bbox_to_anchor=(1.04, 1), loc="upper left") fig.savefig(Path(save_dir), dpi=250) plt.close() ================================================ FILE: RStask/ObjectDetection/utils/plots.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ Plotting utils """ import math import os from copy import copy from pathlib import Path from urllib.error import URLError import cv2 import matplotlib import matplotlib.pyplot as plt import numpy as np import pandas as pd import seaborn as sn import torch from PIL import Image, ImageDraw, ImageFont from RStask.ObjectDetection.utils.general import (CONFIG_DIR, FONT, LOGGER, Timeout, check_font, check_requirements, clip_coords, increment_path, is_ascii, try_except, xywh2xyxy, xyxy2xywh) from RStask.ObjectDetection.utils.metrics import fitness # Settings RANK = int(os.getenv('RANK', -1)) matplotlib.rc('font', **{'size': 11}) matplotlib.use('Agg') # for writing to files only class Colors: # Ultralytics color palette https://ultralytics.com/ def __init__(self): # hex = matplotlib.colors.TABLEAU_COLORS.values() hex = ('FF3838', 'FF9D97', 'FF701F', 'FFB21D', 'CFD231', '48F90A', '92CC17', '3DDB86', '1A9334', '00D4BB', '2C99A8', '00C2FF', '344593', '6473FF', '0018EC', '8438FF', '520085', 'CB38FF', 'FF95C8', 'FF37C7') self.palette = [self.hex2rgb('#' + c) for c in hex] self.n = len(self.palette) def __call__(self, i, bgr=False): c = self.palette[int(i) % self.n] return (c[2], c[1], c[0]) if bgr else c @staticmethod def hex2rgb(h): # rgb order (PIL) return tuple(int(h[1 + i:1 + i + 2], 16) for i in (0, 2, 4)) colors = Colors() # create instance for 'from utils.plots import colors' def check_pil_font(font=FONT, size=10): # Return a PIL TrueType Font, downloading to CONFIG_DIR if necessary font = Path(font) font = font if font.exists() else (CONFIG_DIR / font.name) try: return ImageFont.truetype(str(font) if font.exists() else font.name, size) except Exception: # download if missing try: check_font(font) return ImageFont.truetype(str(font), size) except TypeError: check_requirements('Pillow>=8.4.0') # known issue https://github.com/ultralytics/yolov5/issues/5374 except URLError: # not online return ImageFont.load_default() class Annotator: # YOLOv5 Annotator for train/val mosaics and jpgs and detect/hub inference annotations def __init__(self, im, line_width=None, font_size=None, font='Arial.ttf', pil=False, example='abc'): assert im.data.contiguous, 'Image not contiguous. Apply np.ascontiguousarray(im) to Annotator() input images.' non_ascii = not is_ascii(example) # non-latin labels, i.e. asian, arabic, cyrillic self.pil = pil or non_ascii if self.pil: # use PIL self.im = im if isinstance(im, Image.Image) else Image.fromarray(im) self.draw = ImageDraw.Draw(self.im) self.font = check_pil_font(font='Arial.Unicode.ttf' if non_ascii else font, size=font_size or max(round(sum(self.im.size) / 2 * 0.035), 12)) else: # use cv2 self.im = im self.lw = line_width or max(round(sum(im.shape) / 2 * 0.003), 2) # line width def box_label(self, box, label='', color=(128, 128, 128), txt_color=(255, 255, 255)): # Add one xyxy box to image with label if self.pil or not is_ascii(label): self.draw.rectangle(box, width=self.lw, outline=color) # box if label: w, h = self.font.getsize(label) # text width, height outside = box[1] - h >= 0 # label fits outside box self.draw.rectangle( (box[0], box[1] - h if outside else box[1], box[0] + w + 1, box[1] + 1 if outside else box[1] + h + 1), fill=color, ) # self.draw.text((box[0], box[1]), label, fill=txt_color, font=self.font, anchor='ls') # for PIL>8.0 self.draw.text((box[0], box[1] - h if outside else box[1]), label, fill=txt_color, font=self.font) else: # cv2 p1, p2 = (int(box[0]), int(box[1])), (int(box[2]), int(box[3])) cv2.rectangle(self.im, p1, p2, color, thickness=self.lw, lineType=cv2.LINE_AA) if label: tf = max(self.lw - 1, 1) # font thickness w, h = cv2.getTextSize(label, 0, fontScale=self.lw / 3, thickness=tf)[0] # text width, height outside = p1[1] - h - 3 >= 0 # label fits outside box p2 = p1[0] + w, p1[1] - h - 3 if outside else p1[1] + h + 3 cv2.rectangle(self.im, p1, p2, color, -1, cv2.LINE_AA) # filled cv2.putText(self.im, label, (p1[0], p1[1] - 2 if outside else p1[1] + h + 2), 0, self.lw / 3, txt_color, thickness=tf, lineType=cv2.LINE_AA) def rectangle(self, xy, fill=None, outline=None, width=1): # Add rectangle to image (PIL-only) self.draw.rectangle(xy, fill, outline, width) def text(self, xy, text, txt_color=(255, 255, 255)): # Add text to image (PIL-only) w, h = self.font.getsize(text) # text width, height self.draw.text((xy[0], xy[1] - h + 1), text, fill=txt_color, font=self.font) def result(self): # Return annotated image as array return np.asarray(self.im) def feature_visualization(x, module_type, stage, n=32, save_dir=Path('runs/detect/exp')): """ x: Features to be visualized module_type: Module type stage: Module stage within model n: Maximum number of feature maps to plot save_dir: Directory to save results """ if 'Detect' not in module_type: batch, channels, height, width = x.shape # batch, channels, height, width if height > 1 and width > 1: f = save_dir / f"stage{stage}_{module_type.split('.')[-1]}_features.png" # filename blocks = torch.chunk(x[0].cpu(), channels, dim=0) # select batch index 0, block by channels n = min(n, channels) # number of plots fig, ax = plt.subplots(math.ceil(n / 8), 8, tight_layout=True) # 8 rows x n/8 cols ax = ax.ravel() plt.subplots_adjust(wspace=0.05, hspace=0.05) for i in range(n): ax[i].imshow(blocks[i].squeeze()) # cmap='gray' ax[i].axis('off') LOGGER.info(f'Saving {f}... ({n}/{channels})') plt.savefig(f, dpi=300, bbox_inches='tight') plt.close() np.save(str(f.with_suffix('.npy')), x[0].cpu().numpy()) # npy save def hist2d(x, y, n=100): # 2d histogram used in labels.png and evolve.png xedges, yedges = np.linspace(x.min(), x.max(), n), np.linspace(y.min(), y.max(), n) hist, xedges, yedges = np.histogram2d(x, y, (xedges, yedges)) xidx = np.clip(np.digitize(x, xedges) - 1, 0, hist.shape[0] - 1) yidx = np.clip(np.digitize(y, yedges) - 1, 0, hist.shape[1] - 1) return np.log(hist[xidx, yidx]) def butter_lowpass_filtfilt(data, cutoff=1500, fs=50000, order=5): from scipy.signal import butter, filtfilt # https://stackoverflow.com/questions/28536191/how-to-filter-smooth-with-scipy-numpy def butter_lowpass(cutoff, fs, order): nyq = 0.5 * fs normal_cutoff = cutoff / nyq return butter(order, normal_cutoff, btype='low', analog=False) b, a = butter_lowpass(cutoff, fs, order=order) return filtfilt(b, a, data) # forward-backward filter def output_to_target(output): # Convert model output to target format [batch_id, class_id, x, y, w, h, conf] targets = [] for i, o in enumerate(output): for *box, conf, cls in o.cpu().numpy(): targets.append([i, cls, *list(*xyxy2xywh(np.array(box)[None])), conf]) return np.array(targets) def plot_images(images, targets, paths=None, fname='images.jpg', names=None, max_size=1920, max_subplots=16): # Plot image grid with labels if isinstance(images, torch.Tensor): images = images.cpu().float().numpy() if isinstance(targets, torch.Tensor): targets = targets.cpu().numpy() if np.max(images[0]) <= 1: images *= 255 # de-normalise (optional) bs, _, h, w = images.shape # batch size, _, height, width bs = min(bs, max_subplots) # limit plot images ns = np.ceil(bs ** 0.5) # number of subplots (square) # Build Image mosaic = np.full((int(ns * h), int(ns * w), 3), 255, dtype=np.uint8) # init for i, im in enumerate(images): if i == max_subplots: # if last batch has fewer images than we expect break x, y = int(w * (i // ns)), int(h * (i % ns)) # block origin im = im.transpose(1, 2, 0) mosaic[y:y + h, x:x + w, :] = im # Resize (optional) scale = max_size / ns / max(h, w) if scale < 1: h = math.ceil(scale * h) w = math.ceil(scale * w) mosaic = cv2.resize(mosaic, tuple(int(x * ns) for x in (w, h))) # Annotate fs = int((h + w) * ns * 0.01) # font size annotator = Annotator(mosaic, line_width=round(fs / 10), font_size=fs, pil=True, example=names) for i in range(i + 1): x, y = int(w * (i // ns)), int(h * (i % ns)) # block origin annotator.rectangle([x, y, x + w, y + h], None, (255, 255, 255), width=2) # borders if paths: annotator.text((x + 5, y + 5 + h), text=Path(paths[i]).name[:40], txt_color=(220, 220, 220)) # filenames if len(targets) > 0: ti = targets[targets[:, 0] == i] # image targets boxes = xywh2xyxy(ti[:, 2:6]).T classes = ti[:, 1].astype('int') labels = ti.shape[1] == 6 # labels if no conf column conf = None if labels else ti[:, 6] # check for confidence presence (label vs pred) if boxes.shape[1]: if boxes.max() <= 1.01: # if normalized with tolerance 0.01 boxes[[0, 2]] *= w # scale to pixels boxes[[1, 3]] *= h elif scale < 1: # absolute coords need scale if image scales boxes *= scale boxes[[0, 2]] += x boxes[[1, 3]] += y for j, box in enumerate(boxes.T.tolist()): cls = classes[j] color = colors(cls) cls = names[cls] if names else cls if labels or conf[j] > 0.25: # 0.25 conf thresh label = f'{cls}' if labels else f'{cls} {conf[j]:.1f}' annotator.box_label(box, label, color=color) annotator.im.save(fname) # save def plot_lr_scheduler(optimizer, scheduler, epochs=300, save_dir=''): # Plot LR simulating training for full epochs optimizer, scheduler = copy(optimizer), copy(scheduler) # do not modify originals y = [] for _ in range(epochs): scheduler.step() y.append(optimizer.param_groups[0]['lr']) plt.plot(y, '.-', label='LR') plt.xlabel('epoch') plt.ylabel('LR') plt.grid() plt.xlim(0, epochs) plt.ylim(0) plt.savefig(Path(save_dir) / 'LR.png', dpi=200) plt.close() def plot_val_txt(): # from utils.plots import *; plot_val() # Plot val.txt histograms x = np.loadtxt('val.txt', dtype=np.float32) box = xyxy2xywh(x[:, :4]) cx, cy = box[:, 0], box[:, 1] fig, ax = plt.subplots(1, 1, figsize=(6, 6), tight_layout=True) ax.hist2d(cx, cy, bins=600, cmax=10, cmin=0) ax.set_aspect('equal') plt.savefig('hist2d.png', dpi=300) fig, ax = plt.subplots(1, 2, figsize=(12, 6), tight_layout=True) ax[0].hist(cx, bins=600) ax[1].hist(cy, bins=600) plt.savefig('hist1d.png', dpi=200) def plot_targets_txt(): # from utils.plots import *; plot_targets_txt() # Plot targets.txt histograms x = np.loadtxt('targets.txt', dtype=np.float32).T s = ['x targets', 'y targets', 'width targets', 'height targets'] fig, ax = plt.subplots(2, 2, figsize=(8, 8), tight_layout=True) ax = ax.ravel() for i in range(4): ax[i].hist(x[i], bins=100, label=f'{x[i].mean():.3g} +/- {x[i].std():.3g}') ax[i].legend() ax[i].set_title(s[i]) plt.savefig('targets.jpg', dpi=200) def plot_val_study(file='', dir='', x=None): # from utils.plots import *; plot_val_study() # Plot file=study.txt generated by val.py (or plot all study*.txt in dir) save_dir = Path(file).parent if file else Path(dir) plot2 = False # plot additional results if plot2: ax = plt.subplots(2, 4, figsize=(10, 6), tight_layout=True)[1].ravel() fig2, ax2 = plt.subplots(1, 1, figsize=(8, 4), tight_layout=True) # for f in [save_dir / f'study_coco_{x}.txt' for x in ['yolov5n6', 'yolov5s6', 'yolov5m6', 'yolov5l6', 'yolov5x6']]: for f in sorted(save_dir.glob('study*.txt')): y = np.loadtxt(f, dtype=np.float32, usecols=[0, 1, 2, 3, 7, 8, 9], ndmin=2).T x = np.arange(y.shape[1]) if x is None else np.array(x) if plot2: s = ['P', 'R', 'mAP@.5', 'mAP@.5:.95', 't_preprocess (ms/img)', 't_inference (ms/img)', 't_NMS (ms/img)'] for i in range(7): ax[i].plot(x, y[i], '.-', linewidth=2, markersize=8) ax[i].set_title(s[i]) j = y[3].argmax() + 1 ax2.plot(y[5, 1:j], y[3, 1:j] * 1E2, '.-', linewidth=2, markersize=8, label=f.stem.replace('study_coco_', '').replace('yolo', 'YOLO')) ax2.plot(1E3 / np.array([209, 140, 97, 58, 35, 18]), [34.6, 40.5, 43.0, 47.5, 49.7, 51.5], 'k.-', linewidth=2, markersize=8, alpha=.25, label='EfficientDet') ax2.grid(alpha=0.2) ax2.set_yticks(np.arange(20, 60, 5)) ax2.set_xlim(0, 57) ax2.set_ylim(25, 55) ax2.set_xlabel('GPU Speed (ms/img)') ax2.set_ylabel('COCO AP val') ax2.legend(loc='lower right') f = save_dir / 'study.png' print(f'Saving {f}...') plt.savefig(f, dpi=300) @try_except # known issue https://github.com/ultralytics/yolov5/issues/5395 @Timeout(30) # known issue https://github.com/ultralytics/yolov5/issues/5611 def plot_labels(labels, names=(), save_dir=Path('')): # plot dataset labels LOGGER.info(f"Plotting labels to {save_dir / 'labels.jpg'}... ") c, b = labels[:, 0], labels[:, 1:].transpose() # classes, boxes nc = int(c.max() + 1) # number of classes x = pd.DataFrame(b.transpose(), columns=['x', 'y', 'width', 'height']) # seaborn correlogram sn.pairplot(x, corner=True, diag_kind='auto', kind='hist', diag_kws=dict(bins=50), plot_kws=dict(pmax=0.9)) plt.savefig(save_dir / 'labels_correlogram.jpg', dpi=200) plt.close() # matplotlib labels matplotlib.use('svg') # faster ax = plt.subplots(2, 2, figsize=(8, 8), tight_layout=True)[1].ravel() y = ax[0].hist(c, bins=np.linspace(0, nc, nc + 1) - 0.5, rwidth=0.8) try: # color histogram bars by class [y[2].patches[i].set_color([x / 255 for x in colors(i)]) for i in range(nc)] # known issue #3195 except Exception: pass ax[0].set_ylabel('instances') if 0 < len(names) < 30: ax[0].set_xticks(range(len(names))) ax[0].set_xticklabels(names, rotation=90, fontsize=10) else: ax[0].set_xlabel('classes') sn.histplot(x, x='x', y='y', ax=ax[2], bins=50, pmax=0.9) sn.histplot(x, x='width', y='height', ax=ax[3], bins=50, pmax=0.9) # rectangles labels[:, 1:3] = 0.5 # center labels[:, 1:] = xywh2xyxy(labels[:, 1:]) * 2000 img = Image.fromarray(np.ones((2000, 2000, 3), dtype=np.uint8) * 255) for cls, *box in labels[:1000]: ImageDraw.Draw(img).rectangle(box, width=1, outline=colors(cls)) # plot ax[1].imshow(img) ax[1].axis('off') for a in [0, 1, 2, 3]: for s in ['top', 'right', 'left', 'bottom']: ax[a].spines[s].set_visible(False) plt.savefig(save_dir / 'labels.jpg', dpi=200) matplotlib.use('Agg') plt.close() def plot_evolve(evolve_csv='path/to/evolve.csv'): # from utils.plots import *; plot_evolve() # Plot evolve.csv hyp evolution results evolve_csv = Path(evolve_csv) data = pd.read_csv(evolve_csv) keys = [x.strip() for x in data.columns] x = data.values f = fitness(x) j = np.argmax(f) # max fitness index plt.figure(figsize=(10, 12), tight_layout=True) matplotlib.rc('font', **{'size': 8}) print(f'Best results from row {j} of {evolve_csv}:') for i, k in enumerate(keys[7:]): v = x[:, 7 + i] mu = v[j] # best single result plt.subplot(6, 5, i + 1) plt.scatter(v, f, c=hist2d(v, f, 20), cmap='viridis', alpha=.8, edgecolors='none') plt.plot(mu, f.max(), 'k+', markersize=15) plt.title(f'{k} = {mu:.3g}', fontdict={'size': 9}) # limit to 40 characters if i % 5 != 0: plt.yticks([]) print(f'{k:>15}: {mu:.3g}') f = evolve_csv.with_suffix('.png') # filename plt.savefig(f, dpi=200) plt.close() print(f'Saved {f}') def plot_results(file='path/to/results.csv', dir=''): # Plot training results.csv. Usage: from utils.plots import *; plot_results('path/to/results.csv') save_dir = Path(file).parent if file else Path(dir) fig, ax = plt.subplots(2, 5, figsize=(12, 6), tight_layout=True) ax = ax.ravel() files = list(save_dir.glob('results*.csv')) assert len(files), f'No results.csv files found in {save_dir.resolve()}, nothing to plot.' for fi, f in enumerate(files): try: data = pd.read_csv(f) s = [x.strip() for x in data.columns] x = data.values[:, 0] for i, j in enumerate([1, 2, 3, 4, 5, 8, 9, 10, 6, 7]): y = data.values[:, j] # y[y == 0] = np.nan # don't show zero values ax[i].plot(x, y, marker='.', label=f.stem, linewidth=2, markersize=8) ax[i].set_title(s[j], fontsize=12) # if j in [8, 9, 10]: # share train and val loss y axes # ax[i].get_shared_y_axes().join(ax[i], ax[i - 5]) except Exception as e: LOGGER.info(f'Warning: Plotting error for {f}: {e}') ax[1].legend() fig.savefig(save_dir / 'results.png', dpi=200) plt.close() def profile_idetection(start=0, stop=0, labels=(), save_dir=''): # Plot iDetection '*.txt' per-image logs. from utils.plots import *; profile_idetection() ax = plt.subplots(2, 4, figsize=(12, 6), tight_layout=True)[1].ravel() s = ['Images', 'Free Storage (GB)', 'RAM Usage (GB)', 'Battery', 'dt_raw (ms)', 'dt_smooth (ms)', 'real-world FPS'] files = list(Path(save_dir).glob('frames*.txt')) for fi, f in enumerate(files): try: results = np.loadtxt(f, ndmin=2).T[:, 90:-30] # clip first and last rows n = results.shape[1] # number of rows x = np.arange(start, min(stop, n) if stop else n) results = results[:, x] t = (results[0] - results[0].min()) # set t0=0s results[0] = x for i, a in enumerate(ax): if i < len(results): label = labels[fi] if len(labels) else f.stem.replace('frames_', '') a.plot(t, results[i], marker='.', label=label, linewidth=1, markersize=5) a.set_title(s[i]) a.set_xlabel('time (s)') # if fi == len(files) - 1: # a.set_ylim(bottom=0) for side in ['top', 'right']: a.spines[side].set_visible(False) else: a.remove() except Exception as e: print(f'Warning: Plotting error for {f}; {e}') ax[1].legend() plt.savefig(Path(save_dir) / 'idetection_profile.png', dpi=200) def save_one_box(xyxy, im, file=Path('im.jpg'), gain=1.02, pad=10, square=False, BGR=False, save=True): # Save image crop as {file} with crop size multiple {gain} and {pad} pixels. Save and/or return crop xyxy = torch.tensor(xyxy).view(-1, 4) b = xyxy2xywh(xyxy) # boxes if square: b[:, 2:] = b[:, 2:].max(1)[0].unsqueeze(1) # attempt rectangle to square b[:, 2:] = b[:, 2:] * gain + pad # box wh * gain + pad xyxy = xywh2xyxy(b).long() clip_coords(xyxy, im.shape) crop = im[int(xyxy[0, 1]):int(xyxy[0, 3]), int(xyxy[0, 0]):int(xyxy[0, 2]), ::(1 if BGR else -1)] if save: file.parent.mkdir(parents=True, exist_ok=True) # make directory f = str(increment_path(file).with_suffix('.jpg')) # cv2.imwrite(f, crop) # https://github.com/ultralytics/yolov5/issues/7007 chroma subsampling issue Image.fromarray(cv2.cvtColor(crop, cv2.COLOR_BGR2RGB)).save(f, quality=95, subsampling=0) return crop ================================================ FILE: RStask/ObjectDetection/utils/torch_utils.py ================================================ # YOLOv5 🚀 by Ultralytics, GPL-3.0 license """ PyTorch utils """ import math import os import platform import subprocess import time import warnings from contextlib import contextmanager from copy import deepcopy from pathlib import Path import torch import torch.distributed as dist import torch.nn as nn import torch.nn.functional as F from RStask.ObjectDetection.utils.general import LOGGER, file_update_date, git_describe try: import thop # for FLOPs computation except ImportError: thop = None # Suppress PyTorch warnings warnings.filterwarnings('ignore', message='User provided device_type of \'cuda\', but CUDA is not available. Disabling') @contextmanager def torch_distributed_zero_first(local_rank: int): # Decorator to make all processes in distributed training wait for each local_master to do something if local_rank not in [-1, 0]: dist.barrier(device_ids=[local_rank]) yield if local_rank == 0: dist.barrier(device_ids=[0]) def device_count(): # Returns number of CUDA devices available. Safe version of torch.cuda.device_count(). Only works on Linux. assert platform.system() == 'Linux', 'device_count() function only works on Linux' try: cmd = 'nvidia-smi -L | wc -l' return int(subprocess.run(cmd, shell=True, capture_output=True, check=True).stdout.decode().split()[-1]) except Exception: return 0 def select_device(device='', batch_size=0, newline=True): # device = 'cpu' or '0' or '0,1,2,3' s = f'YOLOv5 🚀 {git_describe() or file_update_date()} torch {torch.__version__} ' # string device = str(device).strip().lower().replace('cuda:', '') # to string, 'cuda:0' to '0' cpu = device == 'cpu' if cpu: os.environ['CUDA_VISIBLE_DEVICES'] = '-1' # force torch.cuda.is_available() = False elif device: # non-cpu device requested os.environ['CUDA_VISIBLE_DEVICES'] = device # set environment variable - must be before assert is_available() assert torch.cuda.is_available() and torch.cuda.device_count() >= len(device.replace(',', '')), \ f"Invalid CUDA '--device {device}' requested, use '--device cpu' or pass valid CUDA device(s)" cuda = not cpu and torch.cuda.is_available() if cuda: devices = device.split(',') if device else '0' # range(torch.cuda.device_count()) # i.e. 0,1,6,7 n = len(devices) # device count if n > 1 and batch_size > 0: # check batch_size is divisible by device_count assert batch_size % n == 0, f'batch-size {batch_size} not multiple of GPU count {n}' space = ' ' * (len(s) + 1) for i, d in enumerate(devices): p = torch.cuda.get_device_properties(i) s += f"{'' if i == 0 else space}CUDA:{d} ({p.name}, {p.total_memory / (1 << 20):.0f}MiB)\n" # bytes to MB else: s += 'CPU\n' if not newline: s = s.rstrip() LOGGER.info(s.encode().decode('ascii', 'ignore') if platform.system() == 'Windows' else s) # emoji-safe return torch.device('cuda:0' if cuda else 'cpu') def time_sync(): # PyTorch-accurate time if torch.cuda.is_available(): torch.cuda.synchronize() return time.time() def profile(input, ops, n=10, device=None): # YOLOv5 speed/memory/FLOPs profiler # # Usage: # input = torch.randn(16, 3, 640, 640) # m1 = lambda x: x * torch.sigmoid(x) # m2 = nn.SiLU() # profile(input, [m1, m2], n=100) # profile over 100 iterations results = [] device = device or select_device() print(f"{'Params':>12s}{'GFLOPs':>12s}{'GPU_mem (GB)':>14s}{'forward (ms)':>14s}{'backward (ms)':>14s}" f"{'input':>24s}{'output':>24s}") for x in input if isinstance(input, list) else [input]: x = x.to(device) x.requires_grad = True for m in ops if isinstance(ops, list) else [ops]: m = m.to(device) if hasattr(m, 'to') else m # device m = m.half() if hasattr(m, 'half') and isinstance(x, torch.Tensor) and x.dtype is torch.float16 else m tf, tb, t = 0, 0, [0, 0, 0] # dt forward, backward try: flops = thop.profile(m, inputs=(x,), verbose=False)[0] / 1E9 * 2 # GFLOPs except Exception: flops = 0 try: for _ in range(n): t[0] = time_sync() y = m(x) t[1] = time_sync() try: _ = (sum(yi.sum() for yi in y) if isinstance(y, list) else y).sum().backward() t[2] = time_sync() except Exception: # no backward method # print(e) # for debug t[2] = float('nan') tf += (t[1] - t[0]) * 1000 / n # ms per op forward tb += (t[2] - t[1]) * 1000 / n # ms per op backward mem = torch.cuda.memory_reserved() / 1E9 if torch.cuda.is_available() else 0 # (GB) s_in = tuple(x.shape) if isinstance(x, torch.Tensor) else 'list' s_out = tuple(y.shape) if isinstance(y, torch.Tensor) else 'list' p = sum(list(x.numel() for x in m.parameters())) if isinstance(m, nn.Module) else 0 # parameters print(f'{p:12}{flops:12.4g}{mem:>14.3f}{tf:14.4g}{tb:14.4g}{str(s_in):>24s}{str(s_out):>24s}') results.append([p, flops, mem, tf, tb, s_in, s_out]) except Exception as e: print(e) results.append(None) torch.cuda.empty_cache() return results def is_parallel(model): # Returns True if model is of type DP or DDP return type(model) in (nn.parallel.DataParallel, nn.parallel.DistributedDataParallel) def de_parallel(model): # De-parallelize a model: returns single-GPU model if model is of type DP or DDP return model.module if is_parallel(model) else model def initialize_weights(model): for m in model.modules(): t = type(m) if t is nn.Conv2d: pass # nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') elif t is nn.BatchNorm2d: m.eps = 1e-3 m.momentum = 0.03 elif t in [nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU]: m.inplace = True def find_modules(model, mclass=nn.Conv2d): # Finds layer indices matching module class 'mclass' return [i for i, m in enumerate(model.module_list) if isinstance(m, mclass)] def sparsity(model): # Return global model sparsity a, b = 0, 0 for p in model.parameters(): a += p.numel() b += (p == 0).sum() return b / a def prune(model, amount=0.3): # Prune model to requested global sparsity import torch.nn.utils.prune as prune print('Pruning model... ', end='') for name, m in model.named_modules(): if isinstance(m, nn.Conv2d): prune.l1_unstructured(m, name='weight', amount=amount) # prune prune.remove(m, 'weight') # make permanent print(' %.3g global sparsity' % sparsity(model)) def fuse_conv_and_bn(conv, bn): # Fuse Conv2d() and BatchNorm2d() layers https://tehnokv.com/posts/fusing-batchnorm-and-conv/ fusedconv = nn.Conv2d(conv.in_channels, conv.out_channels, kernel_size=conv.kernel_size, stride=conv.stride, padding=conv.padding, groups=conv.groups, bias=True).requires_grad_(False).to(conv.weight.device) # Prepare filters w_conv = conv.weight.clone().view(conv.out_channels, -1) w_bn = torch.diag(bn.weight.div(torch.sqrt(bn.eps + bn.running_var))) fusedconv.weight.copy_(torch.mm(w_bn, w_conv).view(fusedconv.weight.shape)) # Prepare spatial bias b_conv = torch.zeros(conv.weight.size(0), device=conv.weight.device) if conv.bias is None else conv.bias b_bn = bn.bias - bn.weight.mul(bn.running_mean).div(torch.sqrt(bn.running_var + bn.eps)) fusedconv.bias.copy_(torch.mm(w_bn, b_conv.reshape(-1, 1)).reshape(-1) + b_bn) return fusedconv def model_info(model, verbose=False, img_size=640): # Model information. img_size may be int or list, i.e. img_size=640 or img_size=[640, 320] n_p = sum(x.numel() for x in model.parameters()) # number parameters n_g = sum(x.numel() for x in model.parameters() if x.requires_grad) # number gradients if verbose: print(f"{'layer':>5} {'name':>40} {'gradient':>9} {'parameters':>12} {'shape':>20} {'mu':>10} {'sigma':>10}") for i, (name, p) in enumerate(model.named_parameters()): name = name.replace('module_list.', '') print('%5g %40s %9s %12g %20s %10.3g %10.3g' % (i, name, p.requires_grad, p.numel(), list(p.shape), p.mean(), p.std())) try: # FLOPs from thop import profile stride = max(int(model.stride.max()), 32) if hasattr(model, 'stride') else 32 img = torch.zeros((1, model.yaml.get('ch', 3), stride, stride), device=next(model.parameters()).device) # input flops = profile(deepcopy(model), inputs=(img,), verbose=False)[0] / 1E9 * 2 # stride GFLOPs img_size = img_size if isinstance(img_size, list) else [img_size, img_size] # expand if int/float fs = ', %.1f GFLOPs' % (flops * img_size[0] / stride * img_size[1] / stride) # 640x640 GFLOPs except (ImportError, Exception): fs = '' name = Path(model.yaml_file).stem.replace('yolov5', 'YOLOv5') if hasattr(model, 'yaml_file') else 'Model' def scale_img(img, ratio=1.0, same_shape=False, gs=32): # img(16,3,256,416) # Scales img(bs,3,y,x) by ratio constrained to gs-multiple if ratio == 1.0: return img else: h, w = img.shape[2:] s = (int(h * ratio), int(w * ratio)) # new size img = F.interpolate(img, size=s, mode='bilinear', align_corners=False) # resize if not same_shape: # pad/crop img h, w = (math.ceil(x * ratio / gs) * gs for x in (h, w)) return F.pad(img, [0, w - s[1], 0, h - s[0]], value=0.447) # value = imagenet mean def copy_attr(a, b, include=(), exclude=()): # Copy attributes from b to a, options to only include [...] and to exclude [...] for k, v in b.__dict__.items(): if (len(include) and k not in include) or k.startswith('_') or k in exclude: continue else: setattr(a, k, v) class EarlyStopping: # YOLOv5 simple early stopper def __init__(self, patience=30): self.best_fitness = 0.0 # i.e. mAP self.best_epoch = 0 self.patience = patience or float('inf') # epochs to wait after fitness stops improving to stop self.possible_stop = False # possible stop may occur next epoch def __call__(self, epoch, fitness): if fitness >= self.best_fitness: # >= 0 to allow for early zero-fitness stage of training self.best_epoch = epoch self.best_fitness = fitness delta = epoch - self.best_epoch # epochs without improvement self.possible_stop = delta >= (self.patience - 1) # possible stop may occur next epoch stop = delta >= self.patience # stop training if patience exceeded if stop: LOGGER.info(f'Stopping training early as no improvement observed in last {self.patience} epochs. ' f'Best results observed at epoch {self.best_epoch}, best model saved as best.pt.\n' f'To update EarlyStopping(patience={self.patience}) pass a new patience value, ' f'i.e. `python train.py --patience 300` or use `--patience 0` to disable EarlyStopping.') return stop class ModelEMA: """ Updated Exponential Moving Average (EMA) from https://github.com/rwightman/pytorch-image-models Keeps a moving average of everything in the model state_dict (parameters and buffers) For EMA details see https://www.tensorflow.org/api_docs/python/tf/train/ExponentialMovingAverage """ def __init__(self, model, decay=0.9999, tau=2000, updates=0): # Create EMA self.ema = deepcopy(de_parallel(model)).eval() # FP32 EMA # if next(model.parameters()).device.type != 'cpu': # self.ema.half() # FP16 EMA self.updates = updates # number of EMA updates self.decay = lambda x: decay * (1 - math.exp(-x / tau)) # decay exponential ramp (to help early epochs) for p in self.ema.parameters(): p.requires_grad_(False) def update(self, model): # Update EMA parameters with torch.no_grad(): self.updates += 1 d = self.decay(self.updates) msd = de_parallel(model).state_dict() # model state_dict for k, v in self.ema.state_dict().items(): if v.dtype.is_floating_point: v *= d v += (1 - d) * msd[k].detach() def update_attr(self, model, include=(), exclude=('process_group', 'reducer')): # Update EMA attributes copy_attr(self.ema, model, include, exclude) ================================================ FILE: RStask/SceneClassification/ResNetScene.py ================================================ import torch from skimage import io class ResNetAID: def __init__(self, device=None): print("Initializing SceneClassification") from torchvision import models self.model = models.resnet34(pretrained=False, num_classes=30) self.device = device try: trained = torch.load('./checkpoints/Res34_AID_best.pth') except: trained = torch.load('../../checkpoints/Res34_AID_best.pth') self.model.load_state_dict(trained) self.model = self.model.to(device) self.model.eval() self.mean, self.std = torch.tensor([123.675, 116.28, 103.53]).reshape((1, 3, 1, 1)), torch.tensor( [58.395, 57.12, 57.375]).reshape((1, 3, 1, 1)) self.all_dict = {'Bridge': 0, 'Medium Residential': 1, 'Park': 2, 'Stadium': 3, 'Church': 4, 'Dense Residential': 5, 'Farmland': 6, 'River': 7, 'School': 8, 'Sparse Residential': 9, 'Viaduct': 10, 'Beach': 11, 'Forest': 12, 'Baseball Field': 13, 'Desert': 14, 'BareLand': 15, 'Railway Station': 16, 'Center': 17, 'Industrial': 18, 'Meadow': 19, 'Airport': 20, 'Storage Tanks': 21, 'Pond': 22, 'Commercial': 23, 'Resort': 24, 'Parking': 25, 'Port': 26, 'Square': 27, 'Mountain': 28, 'Playground': 29} def inference(self, inputs): image_path = inputs image = torch.from_numpy(io.imread(image_path)) image = (image.permute(2, 0, 1).unsqueeze(0) - self.mean) / self.std with torch.no_grad(): pred = self.model(image.to(self.device)) values, indices = torch.softmax(pred, 1).topk(2, dim=1, largest=True, sorted=True) output_txt = image_path + ' has ' + str( torch.round(values[0][0] * 10000).item() / 100) + '% probability being ' + list(self.all_dict.keys())[ indices[0][0]] + ' and ' + str( torch.round(values[0][1] * 10000).item() / 100) + '% probability being ' + list(self.all_dict.keys())[ indices[0][1]]+'.' print(f"\nProcessed Scene Classification, Input Image: {inputs}, Output Scene: {output_txt}") return output_txt ================================================ FILE: RStask/SceneClassification/__init__.py ================================================ ================================================ FILE: RStask/SceneClassification/test.py ================================================ from RStask.SceneClassification.ResNetScene import ResNetAID as SceneFunction model=SceneFunction() model.inference('/data/haonan.guo/RSChatGPT/test.tif') ================================================ FILE: RStask/__init__.py ================================================ from RStask.EdgeDetection.Canny import Image2Canny as ImageEdgeFunction from RStask.ImageCaptioning.blip import BLIP as CaptionFunction from RStask.LanduseSegmentation.seg_hrnet import HRNet48 as LanduseFunction from RStask.ObjectCounting.Yolocounting import YoloCounting as CountingFuncnction from RStask.ObjectDetection.YOLOv5 import YoloDetection as DetectionFunction from RStask.SceneClassification.ResNetScene import ResNetAID as SceneFunction from RStask.InstanceSegmentation.SwinUpper import SwinInstance as InstanceFunction ================================================ FILE: requirements.txt ================================================ langchain torch torchvision imageio numpy openai opencv-python scikit-image