# This source code is licensed under the license found in the # LICENSE file in the root directory of this source tree. # -------------------------------------------------------- # References: # DeiT: https://github.com/facebookresearch/deit # MoCo v3: https://github.com/facebookresearch/moco-v3 # MAE: https://github.com/facebookresearch/mae # -------------------------------------------------------- import argparse import datetime import json import numpy as np import os import time from pathlib import Path import torch import torch.backends.cudnn as cudnn from torch.utils.tensorboard import SummaryWriter import torchvision.transforms as transforms import timm assert timm.__version__ == "0.3.2" # version check from timm.models.layers import trunc_normal_ import util.misc as misc from util.pos_embed import interpolate_pos_embed from util.misc import NativeScalerWithGradNormCount as NativeScaler from util.crop import RandomResizedCrop from util.datasets import ImageListFolder import models_vit from engine_finetune import train_one_epoch, evaluate def get_args_parser(): parser = argparse.ArgumentParser('Hard Patches Mining for Masked Image Modeling', add_help=False) parser.add_argument('--batch_size', default=512, type=int, help='Batch size per GPU (effective batch size is batch_size * accum_iter * # gpus') parser.add_argument('--epochs', default=90, type=int) parser.add_argument('--accum_iter', default=1, type=int, help='Accumulate gradient iterations (for increasing the effective batch size under memory constraints)') # Model parameters parser.add_argument('--model', default='vit_large_patch16', type=str, metavar='MODEL', help='Name of model to train') # Optimizer parameters parser.add_argument('--weight_decay', type=float, default=0, help='weight decay (default: 0 for linear probe following MoCo v1)') parser.add_argument('--lr', type=float, default=None, metavar='LR', help='learning rate (absolute lr)') parser.add_argument('--blr', type=float, default=0.1, metavar='LR', help='base learning rate: absolute_lr = base_lr * total_batch_size / 256') parser.add_argument('--min_lr', type=float, default=0., metavar='LR', help='lower lr bound for cyclic schedulers that hit 0') parser.add_argument('--warmup_epochs', type=int, default=10, metavar='N', help='epochs to warmup LR') # * Finetuning params parser.add_argument('--finetune', default='', help='finetune from checkpoint') parser.add_argument('--global_pool', action='store_true') parser.set_defaults(global_pool=False) parser.add_argument('--cls_token', action='store_false', dest='global_pool', help='Use class token instead of global pool for classification') # Dataset parameters parser.add_argument('--data_path', default='/datasets01/imagenet_full_size/061417/', type=str, help='dataset path') parser.add_argument('--dataloader_type', type=str, default='nori', help="""dataloader type, folder, nori, dpflow..""") parser.add_argument('--nb_classes', default=1000, type=int, help='number of the classification types') parser.add_argument('--output_dir', default='./output_dir', help='path where to save, empty for no saving') parser.add_argument('--log_dir', default='./output_dir', help='path where to tensorboard log') parser.add_argument('--device', default='cuda', help='device to use for training / testing') parser.add_argument('--seed', default=0, type=int) parser.add_argument('--resume', default='', help='resume from checkpoint') parser.add_argument('--start_epoch', default=0, type=int, metavar='N', help='start epoch') parser.add_argument('--eval', action='store_true', help='Perform evaluation only') parser.add_argument('--dist_eval', action='store_true', default=False, help='Enabling distributed evaluation (recommended during training for faster monitor') parser.add_argument('--num_workers', default=10, type=int) parser.add_argument('--pin_mem', action='store_true', help='Pin CPU memory in DataLoader for more efficient (sometimes) transfer to GPU.') parser.add_argument('--no_pin_mem', action='store_false', dest='pin_mem') parser.set_defaults(pin_mem=True) # distributed training parameters parser.add_argument('--world_size', default=1, type=int, help='number of distributed processes') parser.add_argument('--local_rank', default=-1, type=int) parser.add_argument('--dist_on_itp', action='store_true') parser.add_argument('--dist_url', default='env://', help='url used to set up distributed training') return parser def main(args): misc.init_distributed_mode(args) print('job dir: {}'.format(os.path.dirname(os.path.realpath(__file__)))) print("{}".format(args).replace(', ', ',\n')) device = torch.device(args.device) # fix the seed for reproducibility seed = args.seed + misc.get_rank() torch.manual_seed(seed) np.random.seed(seed) cudnn.benchmark = True # linear probe: weak augmentation transform_train = transforms.Compose([ RandomResizedCrop(224, interpolation=3), transforms.RandomHorizontalFlip(), transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])]) transform_val = transforms.Compose([ transforms.Resize(256, interpolation=3), transforms.CenterCrop(224), transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])]) dataset_train = ImageListFolder(os.path.join(args.data_path, 'train'), transform=transform_train, ann_file=os.path.join(args.data_path, 'train.txt')) print(dataset_train) num_tasks = misc.get_world_size() global_rank = misc.get_rank() sampler_train = torch.utils.data.DistributedSampler( dataset_train, num_replicas=num_tasks, rank=global_rank, shuffle=True ) print("Sampler_train = %s" % str(sampler_train)) data_loader_train = torch.utils.data.DataLoader( dataset_train, sampler=sampler_train, batch_size=args.batch_size, num_workers=args.num_workers, pin_memory=args.pin_mem, drop_last=True, ) dataset_val = ImageListFolder(os.path.join(args.data_path, 'val'), transform=transform_val, ann_file=os.path.join(args.data_path, 'val.txt')) print(dataset_val) sampler_val = torch.utils.data.DistributedSampler( dataset_val, num_replicas=num_tasks, rank=global_rank, shuffle=False ) print("Sampler_val = %s" % str(sampler_val)) data_loader_val = torch.utils.data.DataLoader( dataset_val, sampler=sampler_val, batch_size=args.batch_size, num_workers=args.num_workers, pin_memory=args.pin_mem, drop_last=True, ) if global_rank == 0 and args.log_dir is not None and not args.eval: os.makedirs(args.log_dir, exist_ok=True) log_writer = SummaryWriter(log_dir=args.log_dir) else: log_writer = None model = models_vit.__dict__[args.model]( num_classes=args.nb_classes, global_pool=args.global_pool, ) if args.finetune and not args.eval: checkpoint = torch.load(args.finetune, map_location='cpu') print("Load pre-trained checkpoint from: %s" % args.finetune) if 'state_dict' in checkpoint.keys(): state_dict = checkpoint['state_dict'] else: state_dict = checkpoint checkpoint_model = {} for name, param in state_dict.items(): if name.startswith('module.'): checkpoint_model[name[7:]] = param else: checkpoint_model[name] = param state_dict = model.state_dict() for k in ['head.weight', 'head.bias']: if k in checkpoint_model and checkpoint_model[k].shape != state_dict[k].shape: print(f"Removing key {k} from pretrained checkpoint") del checkpoint_model[k] # interpolate position embedding interpolate_pos_embed(model, checkpoint_model) # load pre-trained model missing_keys, unexpected_keys = model.load_state_dict(checkpoint_model, strict=False) print('missing keys:', missing_keys) print('unexpected keys:', unexpected_keys) if args.global_pool: assert set(missing_keys) == {'head.weight', 'head.bias', 'fc_norm.weight', 'fc_norm.bias'} else: assert set(missing_keys) == {'head.weight', 'head.bias'} # manually initialize fc layer: following MoCo v3 trunc_normal_(model.head.weight, std=0.01) # for linear prob only # hack: revise model's head with BN model.head = torch.nn.Sequential(torch.nn.BatchNorm1d(model.head.in_features, affine=False, eps=1e-6), model.head) # freeze all but the head for _, p in model.named_parameters(): p.requires_grad = False for _, p in model.head.named_parameters(): p.requires_grad = True model.to(device) model_without_ddp = model n_parameters = sum(p.numel() for p in model.parameters() if p.requires_grad) print("Model = %s" % str(model_without_ddp)) print('number of params (M): %.2f' % (n_parameters / 1.e6)) eff_batch_size = args.batch_size * args.accum_iter * misc.get_world_size() if args.lr is None: # only base_lr is specified args.lr = args.blr * eff_batch_size / 256 print("base lr: %.2e" % (args.lr * 256 / eff_batch_size)) print("actual lr: %.2e" % args.lr) print("accumulate grad iterations: %d" % args.accum_iter) print("effective batch size: %d" % eff_batch_size) if args.distributed: model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.gpu]) model_without_ddp = model.module # optimizer = LARS(model_without_ddp.head.parameters(), lr=args.lr, weight_decay=args.weight_decay) optimizer = torch.optim.SGD( model_without_ddp.head.parameters(), lr=args.lr, momentum=0.9, weight_decay=0, ) print(optimizer) loss_scaler = NativeScaler() criterion = torch.nn.CrossEntropyLoss() print("criterion = %s" % str(criterion)) # misc.load_model(args=args, model_without_ddp=model_without_ddp, optimizer=optimizer, loss_scaler=loss_scaler) if args.eval: test_stats = evaluate(data_loader_val, model, device) print(f"Accuracy of the network on the {len(dataset_val)} test images: {test_stats['acc1']:.1f}%") exit(0) print(f"Start training for {args.epochs} epochs") start_time = time.time() max_accuracy = 0.0 for epoch in range(args.start_epoch, args.epochs): if args.distributed: data_loader_train.sampler.set_epoch(epoch) train_stats = train_one_epoch( model, criterion, data_loader_train, optimizer, device, epoch, loss_scaler, max_norm=None, log_writer=log_writer, args=args ) if args.output_dir: misc.save_model( args=args, model=model, model_without_ddp=model_without_ddp, optimizer=optimizer, loss_scaler=loss_scaler, epoch=epoch) test_stats = evaluate(data_loader_val, model, device) print(f"Accuracy of the network on the {len(dataset_val)} test images: {test_stats['acc1']:.1f}%") max_accuracy = max(max_accuracy, test_stats["acc1"]) print(f'Max accuracy: {max_accuracy:.2f}%') print(f'Pre-trained from: {args.finetune}') if log_writer is not None: log_writer.add_scalar('perf/test_acc1', test_stats['acc1'], epoch) log_writer.add_scalar('perf/test_acc5', test_stats['acc5'], epoch) log_writer.add_scalar('perf/test_loss', test_stats['loss'], epoch) log_stats = {**{f'train_{k}': v for k, v in train_stats.items()}, **{f'test_{k}': v for k, v in test_stats.items()}, 'epoch': epoch, 'n_parameters': n_parameters} if args.output_dir and misc.is_main_process(): if log_writer is not None: log_writer.flush() with open(os.path.join(args.output_dir, "log.txt"), mode="a", encoding="utf-8") as f: f.write(json.dumps(log_stats) + "\n") total_time = time.time() - start_time total_time_str = str(datetime.timedelta(seconds=int(total_time))) print('Training time {}'.format(total_time_str)) if __name__ == '__main__': args = get_args_parser() args = args.parse_args() if args.output_dir: Path(args.output_dir).mkdir(parents=True, exist_ok=True) main(args)