import argparse import time from models import * from utils.datasets import * from utils.utils import * parser = argparse.ArgumentParser() parser.add_argument('-epochs', type=int, default=100, help='number of epochs') parser.add_argument('-batch_size', type=int, default=16, help='size of each image batch') parser.add_argument('-data_config_path', type=str, default='cfg/coco.data', help='data config file path') parser.add_argument('-cfg', type=str, default='cfg/yolov3.cfg', help='cfg file path') parser.add_argument('-img_size', type=int, default=32 * 13, help='size of each image dimension') parser.add_argument('-resume', default=False, help='resume training flag') parser.add_argument('-multi_scale', default=True, help='train at random img_size 320-608') # ensure memory for 608 size opt = parser.parse_args() print(opt) cuda = torch.cuda.is_available() device = torch.device('cuda:0' if cuda else 'cpu') random.seed(0) np.random.seed(0) torch.manual_seed(0) if cuda: torch.cuda.manual_seed(0) torch.cuda.manual_seed_all(0) torch.backends.cudnn.benchmark = True def main(opt): os.makedirs('weights', exist_ok=True) # Configure run data_config = parse_data_config(opt.data_config_path) num_classes = int(data_config['classes']) if platform == 'darwin': # MacOS (local) train_path = data_config['train'] else: # linux (cloud, i.e. gcp) train_path = '../coco/trainvalno5k.part' # Initialize model model = Darknet(opt.cfg, opt.img_size if opt.multi_scale is False else 608) # Get dataloader dataloader = load_images_and_labels(train_path, batch_size=opt.batch_size, img_size=opt.img_size, augment=True) # Reload saved optimizer state start_epoch = 0 best_loss = float('inf') if opt.resume: checkpoint = torch.load('weights/latest.pt', map_location='cpu') model.load_state_dict(checkpoint['model']) if torch.cuda.device_count() > 1: print('Using ', torch.cuda.device_count(), ' GPUs') model = nn.DataParallel(model) model.to(device).train() # # Transfer learning (train only YOLO layers) # for i, (name, p) in enumerate(model.named_parameters()): # if p.shape[0] != 650: # not YOLO layer # p.requires_grad = False # Set optimizer # optimizer = torch.optim.Adam(filter(lambda p: p.requires_grad, model.parameters())) optimizer = torch.optim.SGD(filter(lambda p: p.requires_grad, model.parameters()), lr=1e-3, momentum=.9, weight_decay=5e-4, nesterov=True) start_epoch = checkpoint['epoch'] + 1 if checkpoint['optimizer'] is not None: optimizer.load_state_dict(checkpoint['optimizer']) best_loss = checkpoint['best_loss'] del checkpoint # current, saved else: # Initialize model with darknet53 weights (optional) if not os.path.isfile('weights/darknet53.conv.74'): os.system('wget https://pjreddie.com/media/files/darknet53.conv.74 -P weights') load_weights(model, 'weights/darknet53.conv.74') if torch.cuda.device_count() > 1: print('Using ', torch.cuda.device_count(), ' GPUs') model = nn.DataParallel(model) model.to(device).train() # Set optimizer # optimizer = torch.optim.Adam(model.parameters(), lr=1e-4, weight_decay=5e-4) optimizer = torch.optim.SGD(model.parameters(), lr=1e-3, momentum=.9, weight_decay=5e-4, nesterov=True) # Set scheduler # scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, milestones=[54, 61], gamma=0.1) model_info(model) t0, t1 = time.time(), time.time() mean_recall, mean_precision = 0, 0 print('%10s' * 16 % ( 'Epoch', 'Batch', 'x', 'y', 'w', 'h', 'conf', 'cls', 'total', 'P', 'R', 'nTargets', 'TP', 'FP', 'FN', 'time')) for epoch in range(opt.epochs): epoch += start_epoch # Multi-Scale YOLO Training if opt.multi_scale: img_size = random.choice(range(10, 20)) * 32 # 320 - 608 pixels dataloader = load_images_and_labels(train_path, batch_size=opt.batch_size, img_size=img_size, augment=True) print('Running Epoch %g at multi_scale img_size %g' % (epoch, img_size)) # Update scheduler (automatic) # scheduler.step() # Update scheduler (manual) at 0, 54, 61 epochs to 1e-3, 1e-4, 1e-5 if epoch < 50: lr = 1e-3 else: lr = 1e-4 for g in optimizer.param_groups: g['lr'] = lr ui = -1 rloss = defaultdict(float) # running loss metrics = torch.zeros(3, num_classes) optimizer.zero_grad() for i, (imgs, targets) in enumerate(dataloader): if sum([len(x) for x in targets]) < 1: # if no targets continue continue # SGD burn-in if (epoch == 0) & (i <= 1000): lr = 1e-3 * (i / 1000) ** 4 for g in optimizer.param_groups: g['lr'] = lr # Compute loss, compute gradient, update parameters loss = model(imgs.to(device), targets, requestPrecision=True) loss.backward() # accumulated_batches = 4 # accumulate gradient for 4 batches before stepping optimizer # if ((i+1) % accumulated_batches == 0) or (i == len(dataloader) - 1): optimizer.step() optimizer.zero_grad() # Compute running epoch-means of tracked metrics ui += 1 metrics += model.losses['metrics'] TP, FP, FN = metrics for key, val in model.losses.items(): rloss[key] = (rloss[key] * ui + val) / (ui + 1) # Precision precision = TP / (TP + FP) k = (TP + FP) > 0 if k.sum() > 0: mean_precision = precision[k].mean() # Recall recall = TP / (TP + FN) k = (TP + FN) > 0 if k.sum() > 0: mean_recall = recall[k].mean() s = ('%10s%10s' + '%10.3g' * 14) % ( '%g/%g' % (epoch, opt.epochs - 1), '%g/%g' % (i, len(dataloader) - 1), rloss['x'], rloss['y'], rloss['w'], rloss['h'], rloss['conf'], rloss['cls'], rloss['loss'], mean_precision, mean_recall, model.losses['nT'], model.losses['TP'], model.losses['FP'], model.losses['FN'], time.time() - t1) t1 = time.time() print(s) # Write epoch results with open('results.txt', 'a') as file: file.write(s + '\n') # Update best loss loss_per_target = rloss['loss'] / rloss['nT'] if loss_per_target < best_loss: best_loss = loss_per_target # Save latest checkpoint checkpoint = {'epoch': epoch, 'best_loss': best_loss, 'model': model.state_dict(), 'optimizer': optimizer.state_dict()} torch.save(checkpoint, 'weights/latest.pt') # Save best checkpoint if best_loss == loss_per_target: os.system('cp weights/latest.pt weights/best.pt') # Save backup weights every 5 epochs if (epoch > 0) & (epoch % 5 == 0): os.system('cp weights/latest.pt weights/backup' + str(epoch) + '.pt') # Save final model dt = time.time() - t0 print('Finished %g epochs in %.2fs (%.2fs/epoch)' % (epoch, dt, dt / (epoch + 1))) if __name__ == '__main__': torch.cuda.empty_cache() main(opt) torch.cuda.empty_cache()