SYMBOL INDEX (295 symbols across 22 files)

FILE: SSC_train.py
  function get_args (line 28) | def get_args():
  function main (line 92) | def main():
  function start (line 112) | def start(local_rank, args):

FILE: datasets/carla_dataset.py
  class CarlaDataset (line 21) | class CarlaDataset(Dataset):
    method __init__ (line 25) | def __init__(self, directory,
    method __len__ (line 104) | def __len__(self):
    method collate_fn (line 107) | def collate_fn(self, data):
    method points_to_voxels (line 113) | def points_to_voxels(self, voxel_grid, points, t_i):
    method get_pose (line 129) | def get_pose(self, idx):
    method __getitem__ (line 135) | def __getitem__(self, idx):
    method find_horizon (line 193) | def find_horizon(self, idx):

FILE: datasets/data.py
  function get_data_id (line 11) | def get_data_id(args):
  function get_class_weights (line 14) | def get_class_weights(freq):
  function get_data (line 23) | def get_data(args):

FILE: layers/Ablation/wo_diffusion.py
  class wo_diff (line 7) | class wo_diff(torch.nn.Module):
    method __init__ (line 8) | def __init__(self, args, multi_criterion) -> None:
    method device (line 22) | def device(self):
    method forward (line 25) | def forward(self, x, input_ten):
    method sample (line 31) | def sample(self, x):

FILE: layers/Latent_Level/stage1/model.py
  function conv3x3x3 (line 10) | def conv3x3x3(in_planes, out_planes, stride=1):
  function conv1x3x3 (line 13) | def conv1x3x3(in_planes, out_planes, stride=1):
  function conv1x1x3 (line 16) | def conv1x1x3(in_planes, out_planes, stride=1):
  function conv1x3x1 (line 19) | def conv1x3x1(in_planes, out_planes, stride=1):
  function conv3x1x1 (line 22) | def conv3x1x1(in_planes, out_planes, stride=1):
  function conv3x1x3 (line 25) | def conv3x1x3(in_planes, out_planes, stride=1):
  function conv1x1 (line 28) | def conv1x1(in_planes, out_planes, stride=1):
  class Asymmetric_Residual_Block (line 32) | class Asymmetric_Residual_Block(nn.Module):
    method __init__ (line 33) | def __init__(self, in_filters, out_filters):
    method forward (line 60) | def forward(self, x):
  class DownBlock (line 81) | class DownBlock(nn.Module):
    method __init__ (line 82) | def __init__(self, in_filters, out_filters, pooling=True, drop_out=Tru...
    method forward (line 93) | def forward(self, x):
  class UpBlock (line 102) | class UpBlock(nn.Module):
    method __init__ (line 103) | def __init__(self, in_filters, out_filters, height_pooling):
    method forward (line 135) | def forward(self, x, skip=False):
  class DDCM (line 159) | class DDCM(nn.Module):
    method __init__ (line 160) | def __init__(self, in_filters, out_filters, kernel_size=(3, 3, 3), str...
    method forward (line 180) | def forward(self, x):
  function l2norm (line 197) | def l2norm(t):
  class Attention (line 200) | class Attention(nn.Module):
    method __init__ (line 201) | def __init__(self, dim, heads = 4, scale = 10):
    method forward (line 208) | def forward(self, x):
  class C_Encoder (line 221) | class C_Encoder(nn.Module):
    method __init__ (line 222) | def __init__(self, args,  nclasses=20, init_size=16, l_size='882', att...
    method forward (line 256) | def forward(self, x, out_conv=True):
  class C_Decoder (line 278) | class C_Decoder(nn.Module):
    method __init__ (line 279) | def __init__(self, args, nclasses=20, init_size=16, l_size='882', atte...
    method forward (line 309) | def forward(self, x, in_conv=True):
  class Completion (line 333) | class Completion(nn.Module):
    method __init__ (line 334) | def __init__(self, args, num_class = 11, init_size=32):
    method forward (line 362) | def forward(self, x):

FILE: layers/Latent_Level/stage1/vector_quantizer.py
  class VectorQuantizer (line 5) | class VectorQuantizer(nn.Module):
    method __init__ (line 7) | def __init__(self,
    method forward (line 19) | def forward(self, z: torch.tensor, point=False) -> torch.tensor: # lat...
    method codebook_to_embedding (line 46) | def codebook_to_embedding(self, encoding_inds, latents_shape): # laten...

FILE: layers/Latent_Level/stage1/vqvae.py
  class vqvae (line 10) | class vqvae(torch.nn.Module):
    method __init__ (line 11) | def __init__(self, args, multi_criterion) -> None:
    method device (line 28) | def device(self):
    method encode (line 31) | def encode(self, x):
    method vector_quantize (line 36) | def vector_quantize(self, latent):
    method coodbook (line 40) | def coodbook(self,quantized_latent_ind, latents_shape):
    method decode (line 44) | def decode(self, quantized_latent):
    method forward (line 49) | def forward(self, x, input_ten):
    method sample (line 58) | def sample(self, x):

FILE: layers/Latent_Level/stage2/Gen_diffusion.py
  function sum_except_batch (line 14) | def sum_except_batch(x, num_dims=1):
  function log_1_min_a (line 18) | def log_1_min_a(a):
  function log_add_exp (line 22) | def log_add_exp(a, b):
  function exists (line 27) | def exists(x):
  function extract (line 31) | def extract(a, t, x_shape):
  function default (line 37) | def default(val, d):
  function log_categorical (line 43) | def log_categorical(log_x_start, log_prob):
  function index_to_log_onehot (line 47) | def index_to_log_onehot(x, num_classes):
  function log_onehot_to_index (line 58) | def log_onehot_to_index(log_x):
  function cosine_beta_schedule (line 62) | def cosine_beta_schedule(timesteps, s = 0.008):
  class latent_diffusion (line 78) | class latent_diffusion(torch.nn.Module):
    method __init__ (line 79) | def __init__(self, args, VAE_DENSE, multi_criterion,
    method device (line 115) | def device(self):
    method multinomial_kl (line 118) | def multinomial_kl(self, log_prob1, log_prob2):
    method q_pred_one_timestep (line 122) | def q_pred_one_timestep(self, log_x_t, t):
    method q_pred (line 135) | def q_pred(self, log_x_start, t):
    method predict_start (line 146) | def predict_start(self, log_x_t, t):
    method q_posterior (line 158) | def q_posterior(self, log_x_start, log_x_t, t):
    method p_pred (line 182) | def p_pred(self, log_x, t):
    method log_sample_categorical (line 187) | def log_sample_categorical(self, logits):
    method q_sample (line 194) | def q_sample(self, log_x_start, t):
    method kl_prior (line 199) | def kl_prior(self, log_x_start):
    method sample_time (line 210) | def sample_time(self, b, device, method='uniform'):
    method forward (line 233) | def forward(self, x, input_data):
    method sample (line 287) | def sample(self, x):

FILE: layers/Latent_Level/stage2/gen_denoise.py
  function conv3x3x3 (line 11) | def conv3x3x3(in_planes, out_planes, stride=1):
  function conv1x3x3 (line 14) | def conv1x3x3(in_planes, out_planes, stride=1):
  function conv1x1x3 (line 18) | def conv1x1x3(in_planes, out_planes, stride=1):
  function conv1x3x1 (line 22) | def conv1x3x1(in_planes, out_planes, stride=1):
  function conv3x1x1 (line 26) | def conv3x1x1(in_planes, out_planes, stride=1):
  function conv3x1x3 (line 30) | def conv3x1x3(in_planes, out_planes, stride=1):
  function conv1x1 (line 34) | def conv1x1(in_planes, out_planes, stride=1):
  class Asymmetric_Residual_Block (line 38) | class Asymmetric_Residual_Block(nn.Module):
    method __init__ (line 39) | def __init__(self, in_filters, out_filters, time_filters=128):
    method forward (line 64) | def forward(self, x, t):
  class DDCM (line 91) | class DDCM(nn.Module):
    method __init__ (line 92) | def __init__(self, in_filters, out_filters, kernel_size=(3, 3, 3), str...
    method forward (line 106) | def forward(self, x):
  function l2norm (line 124) | def l2norm(t):
  class Attention (line 127) | class Attention(nn.Module):
    method __init__ (line 128) | def __init__(self, dim, heads = 4, scale = 10):
    method forward (line 135) | def forward(self, x):
  class Cross_Attention (line 148) | class Cross_Attention(nn.Module):
    method __init__ (line 149) | def __init__(self, dim, heads = 4, scale = 10):
    method forward (line 159) | def forward(self, x, cond_x):
  class DownBlock (line 175) | class DownBlock(nn.Module):
    method __init__ (line 176) | def __init__(self, in_filters, out_filters, dropout_rate, kernel_size=...
    method forward (line 193) | def forward(self, x, t):
  class UpBlock (line 201) | class UpBlock(nn.Module):
    method __init__ (line 202) | def __init__(self, in_filters, out_filters, height_pooling, time_filte...
    method forward (line 231) | def forward(self, x, residual, t):
  function timestep_embedding (line 258) | def timestep_embedding(timesteps, dim, max_period=10000, repeat_only=Fal...
  class Denoise (line 280) | class Denoise(nn.Module):
    method __init__ (line 281) | def __init__(self, args, num_class = 11, init_size=32, discrete=True):
    method forward (line 324) | def forward(self, x, t):

FILE: layers/Voxel_Level/Con_Diffusion.py
  function sum_except_batch (line 14) | def sum_except_batch(x, num_dims=1):
  function log_1_min_a (line 18) | def log_1_min_a(a):
  function log_add_exp (line 22) | def log_add_exp(a, b):
  function exists (line 27) | def exists(x):
  function extract (line 31) | def extract(a, t, x_shape):
  function default (line 37) | def default(val, d):
  function log_categorical (line 43) | def log_categorical(log_x_start, log_prob):
  function index_to_log_onehot (line 47) | def index_to_log_onehot(x, num_classes):
  function log_onehot_to_index (line 57) | def log_onehot_to_index(log_x):
  function cosine_beta_schedule (line 61) | def cosine_beta_schedule(timesteps, s = 0.008):
  class Con_Diffusion (line 77) | class Con_Diffusion(torch.nn.Module):
    method __init__ (line 78) | def __init__(self, args, multi_criterion,auxiliary_loss_weight=0.05, a...
    method device (line 117) | def device(self):
    method multinomial_kl (line 120) | def multinomial_kl(self, log_prob1, log_prob2):
    method q_pred_one_timestep (line 124) | def q_pred_one_timestep(self, log_x_t, t):
    method q_pred (line 137) | def q_pred(self, log_x_start, t):
    method predict_start (line 148) | def predict_start(self, log_x_t, t, cond):
    method q_posterior (line 156) | def q_posterior(self, log_x_start, log_x_t, t):
    method p_pred (line 177) | def p_pred(self, log_x, t, cond):
    method log_sample_categorical (line 182) | def log_sample_categorical(self, logits):
    method q_sample (line 189) | def q_sample(self, log_x_start, t):
    method kl_prior (line 194) | def kl_prior(self, log_x_start):
    method sample_time (line 205) | def sample_time(self, b, device, method='uniform'):
    method forward (line 228) | def forward(self, x, voxel_input):
    method sample (line 280) | def sample(self, voxel_input, intermediate=False):

FILE: layers/Voxel_Level/Gen_Diffusion.py
  function sum_except_batch (line 14) | def sum_except_batch(x, num_dims=1):
  function log_1_min_a (line 18) | def log_1_min_a(a):
  function log_add_exp (line 22) | def log_add_exp(a, b):
  function exists (line 27) | def exists(x):
  function extract (line 31) | def extract(a, t, x_shape):
  function default (line 37) | def default(val, d):
  function log_categorical (line 43) | def log_categorical(log_x_start, log_prob):
  function index_to_log_onehot (line 47) | def index_to_log_onehot(x, num_classes):
  function log_onehot_to_index (line 57) | def log_onehot_to_index(log_x):
  function cosine_beta_schedule (line 61) | def cosine_beta_schedule(timesteps, s = 0.008):
  class Diffusion (line 77) | class Diffusion(torch.nn.Module):
    method __init__ (line 78) | def __init__(self, args, multi_criterion,auxiliary_loss_weight=0.05, a...
    method device (line 114) | def device(self):
    method multinomial_kl (line 117) | def multinomial_kl(self, log_prob1, log_prob2):
    method q_pred_one_timestep (line 121) | def q_pred_one_timestep(self, log_x_t, t):
    method q_pred (line 134) | def q_pred(self, log_x_start, t):
    method predict_start (line 145) | def predict_start(self, log_x_t, t):
    method q_posterior (line 153) | def q_posterior(self, log_x_start, log_x_t, t):
    method p_pred (line 174) | def p_pred(self, log_x, t):
    method log_sample_categorical (line 179) | def log_sample_categorical(self, logits):
    method q_sample (line 186) | def q_sample(self, log_x_start, t):
    method kl_prior (line 191) | def kl_prior(self, log_x_start):
    method sample_time (line 202) | def sample_time(self, b, device, method='uniform'):
    method forward (line 225) | def forward(self, x, voxel_input):
    method sample (line 277) | def sample(self, voxel_input):

FILE: layers/Voxel_Level/denoise.py
  function conv3x3x3 (line 11) | def conv3x3x3(in_planes, out_planes, stride=1):
  function conv1x3x3 (line 14) | def conv1x3x3(in_planes, out_planes, stride=1):
  function conv1x1x3 (line 18) | def conv1x1x3(in_planes, out_planes, stride=1):
  function conv1x3x1 (line 22) | def conv1x3x1(in_planes, out_planes, stride=1):
  function conv3x1x1 (line 26) | def conv3x1x1(in_planes, out_planes, stride=1):
  function conv3x1x3 (line 30) | def conv3x1x3(in_planes, out_planes, stride=1):
  function conv1x1 (line 34) | def conv1x1(in_planes, out_planes, stride=1):
  class Asymmetric_Residual_Block (line 38) | class Asymmetric_Residual_Block(nn.Module):
    method __init__ (line 39) | def __init__(self, in_filters, out_filters, time_filters=32*4):
    method forward (line 71) | def forward(self, x, t):
  class DDCM (line 98) | class DDCM(nn.Module):
    method __init__ (line 99) | def __init__(self, in_filters, out_filters, kernel_size=(3, 3, 3), str...
    method forward (line 118) | def forward(self, x):
  function l2norm (line 136) | def l2norm(t):
  class Attention (line 139) | class Attention(nn.Module):
    method __init__ (line 140) | def __init__(self, dim, heads = 4, scale = 10):
    method forward (line 147) | def forward(self, x):
  class Cross_Attention (line 160) | class Cross_Attention(nn.Module):
    method __init__ (line 161) | def __init__(self, dim, heads = 4, scale = 10):
    method forward (line 171) | def forward(self, x, cond_x):
  class DownBlock (line 187) | class DownBlock(nn.Module):
    method __init__ (line 188) | def __init__(self, in_filters, out_filters, time_filters=32*4, kernel_...
    method forward (line 201) | def forward(self, x, t):
  class UpBlock (line 209) | class UpBlock(nn.Module):
    method __init__ (line 210) | def __init__(self, in_filters, out_filters, height_pooling, time_filte...
    method forward (line 245) | def forward(self, x, residual, t):
  function timestep_embedding (line 272) | def timestep_embedding(timesteps, dim, max_period=10000, repeat_only=Fal...
  class Denoise (line 286) | class Denoise(nn.Module):
    method __init__ (line 287) | def __init__(self, args, num_class = 11, init_size=32, discrete=True):
    method forward (line 322) | def forward(self, x, x_cond, t):

FILE: layers/Voxel_Level/gen_denoise.py
  function conv3x3x3 (line 11) | def conv3x3x3(in_planes, out_planes, stride=1):
  function conv1x3x3 (line 14) | def conv1x3x3(in_planes, out_planes, stride=1):
  function conv1x1x3 (line 18) | def conv1x1x3(in_planes, out_planes, stride=1):
  function conv1x3x1 (line 22) | def conv1x3x1(in_planes, out_planes, stride=1):
  function conv3x1x1 (line 26) | def conv3x1x1(in_planes, out_planes, stride=1):
  function conv3x1x3 (line 30) | def conv3x1x3(in_planes, out_planes, stride=1):
  function conv1x1 (line 34) | def conv1x1(in_planes, out_planes, stride=1):
  class Asymmetric_Residual_Block (line 38) | class Asymmetric_Residual_Block(nn.Module):
    method __init__ (line 39) | def __init__(self, in_filters, out_filters, time_filters=128):
    method forward (line 70) | def forward(self, x, t):
  class DDCM (line 97) | class DDCM(nn.Module):
    method __init__ (line 98) | def __init__(self, in_filters, out_filters, kernel_size=(3, 3, 3), str...
    method forward (line 115) | def forward(self, x):
  function l2norm (line 132) | def l2norm(t):
  class Attention (line 135) | class Attention(nn.Module):
    method __init__ (line 136) | def __init__(self, dim, heads = 4, scale = 10):
    method forward (line 143) | def forward(self, x):
  class Cross_Attention (line 156) | class Cross_Attention(nn.Module):
    method __init__ (line 157) | def __init__(self, dim, heads = 4, scale = 10):
    method forward (line 167) | def forward(self, x, cond_x):
  class DownBlock (line 183) | class DownBlock(nn.Module):
    method __init__ (line 184) | def __init__(self, in_filters, out_filters, time_filters, kernel_size=...
    method forward (line 200) | def forward(self, x, t):
  class UpBlock (line 208) | class UpBlock(nn.Module):
    method __init__ (line 209) | def __init__(self, in_filters, out_filters, height_pooling, time_filte...
    method forward (line 244) | def forward(self, x, residual, t):
  function timestep_embedding (line 271) | def timestep_embedding(timesteps, dim, max_period=10000, repeat_only=Fal...
  class Denoise (line 293) | class Denoise(nn.Module):
    method __init__ (line 294) | def __init__(self, args, num_class = 11, init_size=32, discrete=True):
    method forward (line 329) | def forward(self, x, t):

FILE: simple_visualize.py
  function load_config (line 7) | def load_config(yaml_path):
  function load_pointcloud (line 12) | def load_pointcloud(filepath, learning_map, color_map):
  function main (line 29) | def main():

FILE: train.py
  class Experiment (line 19) | class Experiment(object):
    method __init__ (line 22) | def __init__(self, args, model, optimizer, scheduler_iter, scheduler_e...
    method run (line 65) | def run(self, epochs):
    method train_fn (line 112) | def train_fn(self, epoch):
    method eval_fn (line 142) | def eval_fn(self, epoch):
    method sample (line 162) | def sample(self):
    method resume (line 215) | def resume(self):
    method log_metrics (line 246) | def log_metrics(self, dict, type):
    method save_metrics (line 254) | def save_metrics(self):
    method checkpoint_save (line 270) | def checkpoint_save(self, epoch):
    method checkpoint_load (line 292) | def checkpoint_load(self, resume_path):

FILE: utils/cuda.py
  function find_free_port (line 11) | def find_free_port():
  function launch (line 23) | def launch(fn, n_gpu_per_machine, n_machine=1, machine_rank=0, dist_url=...
  function distributed_worker (line 54) | def distributed_worker(local_rank, fn, world_size, n_gpu_per_machine, ma...
  function set_cuda_vd (line 94) | def set_cuda_vd(gpu_ids, verbose=True):

FILE: utils/dicts.py
  function is_primary (line 13) | def is_primary():
  function get_rank (line 17) | def get_rank():
  function get_local_rank (line 27) | def get_local_rank():
  function synchronize (line 40) | def synchronize():
  function get_world_size (line 55) | def get_world_size():
  function is_distributed (line 65) | def is_distributed():
  function all_reduce (line 70) | def all_reduce(tensor, op=dist.ReduceOp.SUM, async_op=False):
  function all_gather (line 80) | def all_gather(data):
  function reduce_dict (line 115) | def reduce_dict(input_dict, average=True):
  function data_sampler (line 140) | def data_sampler(dataset, shuffle, distributed):
  function clean_dict (line 150) | def clean_dict(d, keys):

FILE: utils/intermediate_vis.py
  class Vis_iter (line 19) | class Vis_iter(object):
    method __init__ (line 22) | def __init__(self, args, model, optimizer, scheduler_iter, scheduler_e...
    method run (line 69) | def run(self, epochs):
    method sample (line 74) | def sample(self):
    method checkpoint_load (line 84) | def checkpoint_load(self, resume_path):

FILE: utils/loss.py
  function dice_coef (line 16) | def dice_coef(y_true, y_pred, smooth=1e-6):
  function dice_coef_multilabel (line 22) | def dice_coef_multilabel(y_true, y_pred, numLabels=11):
  function lovasz_grad (line 33) | def lovasz_grad(gt_sorted):
  function lovasz_softmax (line 50) | def lovasz_softmax(probas, labels, classes='present', per_image=False, i...
  function lovasz_softmax_flat (line 68) | def lovasz_softmax_flat(probas, labels, classes='present'):
  function flatten_probas (line 99) | def flatten_probas(probas, labels, ignore=None):
  function isnan (line 123) | def isnan(x):
  function mean (line 127) | def mean(l, ignore_nan=False, empty=0):

FILE: utils/multistep.py
  class LinearWarmupScheduler (line 5) | class LinearWarmupScheduler(_LRScheduler):
    method __init__ (line 12) | def __init__(self, optimizer, total_epoch, last_epoch=-1):
    method get_lr (line 16) | def get_lr(self):
  function get_optim (line 21) | def get_optim(args, model):

FILE: utils/tables.py
  function get_args_table (line 9) | def get_args_table(args_dict):
  function get_miou_table (line 15) | def get_miou_table(args, label_to_names, miou):
  function get_metric_table (line 21) | def get_metric_table(metric_dict, epochs):
  function create_folders (line 29) | def create_folders(args):
  function inter_vis (line 38) | def inter_vis(args, recons):
  function visualization (line 49) | def visualization(args, recons, input_data, output, invalid, iteration):
  function completion_vis (line 74) | def completion_vis(args, input_p, recons):
  function iou_one_frame (line 82) | def iou_one_frame(pred, target, n_classes=23):
  function get_result (line 94) | def get_result(args, for_mask, output, preds, SSC=True):
  function save_args (line 127) | def save_args(args):
  function print_completion (line 137) | def print_completion(num_correct, num_total, TP, FP, FN):
  function print_result (line 154) | def print_result(args, label_to_names, num_correct, num_total, all_inter...

FILE: visualization.py
  class SpheresApp (line 20) | class SpheresApp:
    method __init__ (line 25) | def __init__(self, opt):
    method _on_menu_before (line 63) | def _on_menu_before(self):
    method _on_menu_quit (line 84) | def _on_menu_quit(self):
    method _on_menu_scene (line 87) | def _on_menu_scene(self):
  function get_voxel (line 108) | def get_voxel(opt):
  function get_input (line 142) | def get_input(opt):
  function main (line 150) | def main(opt):