import argparse import os from pathlib import Path import torch from torch import autocast, Tensor from torch.nn.functional import binary_cross_entropy_with_logits from torch.cuda.amp import GradScaler from tqdm import tqdm from config import parser from einops import rearrange, reduce, repeat from dataloaders.JSRT import build_dataloaders from models.unet_model import Unet from trainers.train_baseline import validate, save from trainers.utils import (TensorboardLogger, compare_configs, seed_everything, crop_batch) def train(config, model, optimizer, train_dl, val_dl, logger, scaler, step): best_val_loss = float('inf') train_losses = [] if config.dataset == "BRATS2D": train_losses_per_class = [] elif config.shared_weights_over_timesteps and config.experiment == 'datasetDM': train_losses_per_timestep = [] pbar = tqdm(total=config.val_freq, desc='Training') while True: for x, y in train_dl: if config.shared_weights_over_timesteps and config.experiment == 'datasetDM': y = repeat(y, 'b c h w -> (b step) c h w', step=len(model.steps)) if config.augment_at_finetuning: x, y = crop_batch([x, y], config.img_size, config.batch_size) brightness = torch.rand((config.batch_size, 1, 1, 1), device=x.device)*.6 - .3 # random brightness adjustment between [-.3, .3] contrast = torch.rand((config.batch_size, 1, 1, 1), device=x.device)*.6 + .7 # random contrast adjustment between [.7, 1.3] x = (x + brightness) * contrast # apply brightness and contrast x = x.to(config.device) y = y.to(config.device) optimizer.zero_grad() with autocast(device_type=config.device, enabled=config.mixed_precision): pred = model(x) # cross entropy loss #loss = - ((y * torch.log(torch.sigmoid(pred)) + (1 - y) * torch.log(1 - torch.sigmoid(pred)))).mean() if config.dataset == "BRATS2D": weights = repeat(torch.Tensor(config.loss_weights).to(config.device), 'c -> b c h w', b=y.shape[0], h=y.shape[2], w=y.shape[3]) else: weights = None expanded_loss = reduce(binary_cross_entropy_with_logits(pred, y, weight=weights, reduction='none'), 'b c h w -> b c', 'mean') loss = expanded_loss.mean() scaler.scale(loss).backward() optimizer.step() train_losses.append(loss.item()) if config.dataset == "BRATS2D": loss_per_class = expanded_loss.mean(0) train_losses_per_class.append(loss_per_class.detach().cpu()) pbar.set_description(f'Training loss: {loss.item():.4f} - {loss_per_class[0].item():.4f} - {loss_per_class[1].item():.4f} - {loss_per_class[2].item():.4f} - {loss_per_class[3].item():.4f}') else: pbar.set_description(f'Training loss: {loss.item():.4f}') pbar.update(1) step += 1 if config.unfreeze_weights_at_step == step: for name, param in model.named_parameters(): if name.startswith('downs') or name.startswith('init_conv') or name.startswith('mid_'): param.requires_grad = True if step % config.log_freq == 0 or config.debug: avg_train_loss = sum(train_losses) / len(train_losses) print(f'Step {step} - Train loss: {avg_train_loss:.4f}') logger.log({'train/loss': avg_train_loss}, step=step) if config.dataset == "BRATS2D": avg_train_loss_per_class = torch.stack(train_losses_per_class).mean(0) logger.log({'train_loss/0':avg_train_loss_per_class[0].item()}, step=step) logger.log({'train_loss/1':avg_train_loss_per_class[1].item()}, step=step) logger.log({'train_loss/2':avg_train_loss_per_class[2].item()}, step=step) logger.log({'train_loss/3':avg_train_loss_per_class[3].item()}, step=step) if config.shared_weights_over_timesteps and config.experiment == 'datasetDM': avg_train_loss_per_timestep = torch.stack(train_losses_per_timestep).mean(0) for i, model_step in enumerate(model.steps): logger.log({'train_loss/step_' + str(model_step): avg_train_loss_per_timestep[i].item()}, step=step) if step % config.val_freq == 0 or config.debug: val_results = validate(config, model, val_dl) logger.log(val_results, step=step) if val_results['val/loss'] < best_val_loss and not config.debug: print(f'Step {step} - New best validation loss: ' f'{val_results["val/loss"]:.4f}, saving model ' f'in {config.log_dir}') best_val_loss = val_results['val/loss'] save( model, optimizer, config, config.log_dir / 'best_model.pt', step ) elif val_results['val/loss'] > best_val_loss * 1.5 and config.early_stop: print(f'Step {step} - Validation loss increased by more than 50%') return model if step >= config.max_steps or config.debug: return model def load(config, path): raise NotImplementedError def main(config): os.makedirs(config.log_dir, exist_ok=True) # save config namespace into logdir with open(config.log_dir / 'config.txt', 'w') as f: for k, v in vars(config).items(): if type(v) not in [str, int, float, bool]: f.write(f'{k}: {str(v)}\n') else: f.write(f'{k}: {v}\n') # Random seed seed_everything(config.seed) # Init model and optimizer if config.resume_path is not None: print('Loading model from', config.resume_path) model, optimizer, step = load(config, config.resume_path) else: model = Unet( img_size=config.img_size, dim=config.dim, dim_mults=config.dim_mults, channels=config.channels, out_dim=config.out_channels) state_dict = torch.load(config.global_model_path, map_location='cpu')['model_state_dict'] out = model.load_state_dict(state_dict=state_dict, strict=False) print("Loaded state dict. \n\tMissing keys: {}\n\tUnexpected keys: {}".format(out.missing_keys, out.unexpected_keys)) print('Note that although the state dict of the decoder is loaded, its values are random.') if config.unfreeze_weights_at_step !=0: for name, param in model.named_parameters(): if name.startswith('downs') or name.startswith('init_conv') or name.startswith('mid_'): param.requires_grad = False optimizer = torch.optim.Adam(model.parameters(), lr=config.lr) # , betas=config.adam_betas) step = 0 model.to(config.device) model.train() scaler = GradScaler() # Load data dataloaders = build_dataloaders( config.data_dir, config.img_size, config.batch_size, config.num_workers, n_labelled_images=config.n_labelled_images, ) train_dl = dataloaders['train'] val_dl = dataloaders['val'] print('Train dataset size:', len(train_dl.dataset)) print('Validation dataset size:', len(val_dl.dataset)) # Logger logger = TensorboardLogger(config.log_dir, enabled=not config.debug) train(config, model, optimizer, train_dl, val_dl, logger, scaler, step)