SYMBOL INDEX (60 symbols across 9 files)

FILE: ape_tools/ape_api.py
  function setup_cfg (line 20) | def setup_cfg():
  function ape_inference (line 51) | def ape_inference(input, text_prompt, demo):

FILE: ape_tools/ape_service.py
  function main (line 5) | def main():

FILE: evals/generate_longvideobench.py
  function process_video (line 35) | def process_video(video_path, max_frames_num, fps=1, force_sample=False):
  function extract_audio (line 54) | def extract_audio(video_path, audio_path):
  function chunk_audio (line 58) | def chunk_audio(audio_path, chunk_length_s=30):
  function transcribe_chunk (line 68) | def transcribe_chunk(chunk):
  function get_asr_docs (line 81) | def get_asr_docs(video_path, audio_path):
  function chunk_subtitles (line 97) | def chunk_subtitles(text_list, chunk_size=20):
  function get_ocr_docs (line 120) | def get_ocr_docs(frames):
  function save_frames (line 139) | def save_frames(frames, file_name):
  function get_det_docs (line 148) | def get_det_docs(frames, prompt, file_name):
  function det_preprocess (line 164) | def det_preprocess(det_docs, location, relation, number):
  function llava_inference (line 196) | def llava_inference(qs, video):

FILE: evals/generate_mlvu.py
  function process_video (line 40) | def process_video(video_path, max_frames_num, fps=1, force_sample=False):
  function extract_audio (line 58) | def extract_audio(video_path, audio_path):
  function chunk_audio (line 62) | def chunk_audio(audio_path, chunk_length_s=30):
  function transcribe_chunk (line 72) | def transcribe_chunk(chunk):
  function get_asr_docs (line 85) | def get_asr_docs(video_path, audio_path):
  function get_ocr_docs (line 101) | def get_ocr_docs(frames):
  function save_frames (line 120) | def save_frames(frames, file_name):
  function get_det_docs (line 129) | def get_det_docs(frames, prompt, file_name):
  function det_preprocess (line 145) | def det_preprocess(det_docs, location, relation, number):
  function llava_inference (line 177) | def llava_inference(qs, video):
  function get_prompt2 (line 214) | def get_prompt2(conv):
  class MLVU (line 228) | class MLVU(Dataset):
    method __init__ (line 229) | def __init__(self, data_dir, video_folder):
    method __str__ (line 243) | def __str__(self):
    method __len__ (line 265) | def __len__(self):
    method get_index (line 268) | def get_index(self, bound, fps, max_frame, first_idx=0):
    method qa_template (line 280) | def qa_template(self, data):
    method __getitem__ (line 291) | def __getitem__(self, idx):
  function chunk_string_by_words (line 304) | def chunk_string_by_words(string, words_per_chunk):

FILE: evals/generate_videomme.py
  function process_video (line 31) | def process_video(video_path, max_frames_num, fps=1, force_sample=False):
  function extract_audio (line 49) | def extract_audio(video_path, audio_path):
  function chunk_audio (line 53) | def chunk_audio(audio_path, chunk_length_s=30):
  function transcribe_chunk (line 63) | def transcribe_chunk(chunk):
  function get_asr_docs (line 76) | def get_asr_docs(video_path, audio_path):
  function get_ocr_docs (line 91) | def get_ocr_docs(frames):
  function save_frames (line 110) | def save_frames(frames, file_name):
  function get_det_docs (line 119) | def get_det_docs(frames, prompt, file_name):
  function det_preprocess (line 135) | def det_preprocess(det_docs, location, relation, number):
  function llava_inference (line 164) | def llava_inference(qs, video):

FILE: vidrag_pipeline/tools/filter_keywords.py
  function filter_keywords (line 9) | def filter_keywords(keywords):

FILE: vidrag_pipeline/tools/rag_retriever_dynamic.py
  function text_to_vector (line 9) | def text_to_vector(text, max_length=512):
  function retrieve_documents_with_dynamic (line 15) | def retrieve_documents_with_dynamic(documents, queries, threshold=0.4):

FILE: vidrag_pipeline/tools/scene_graph.py
  function calculate_xmax_ymax (line 3) | def calculate_xmax_ymax(bbox):
  function calculate_spatial_relations (line 9) | def calculate_spatial_relations(bbox1, bbox2):
  function relation_to_text (line 36) | def relation_to_text(source_id, source_label, relation, target_id, targe...
  function generate_scene_graph_description (line 52) | def generate_scene_graph_description(objects, location_des, relation_des...

FILE: vidrag_pipeline/vidrag_pipeline.py
  function process_video (line 34) | def process_video(video_path, max_frames_num, fps=1, force_sample=False):
  function extract_audio (line 53) | def extract_audio(video_path, audio_path):
  function chunk_audio (line 57) | def chunk_audio(audio_path, chunk_length_s=30):
  function transcribe_chunk (line 67) | def transcribe_chunk(chunk):
  function get_asr_docs (line 80) | def get_asr_docs(video_path, audio_path):
  function get_ocr_docs (line 96) | def get_ocr_docs(frames):
  function save_frames (line 115) | def save_frames(frames):
  function get_det_docs (line 124) | def get_det_docs(frames, prompt):
  function det_preprocess (line 140) | def det_preprocess(det_docs, location, relation, number):
  function llava_inference (line 175) | def llava_inference(qs, video):