# -------------------------------------------------------- # BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizers (https://arxiv.org/abs/2208.06366) # Github source: https://github.com/microsoft/unilm/tree/master/beitv2 # Copyright (c) 2022 Microsoft # Licensed under The MIT License [see LICENSE for details] # By Zhiliang Peng # Based on BEiT, timm, DeiT and DINO code bases # https://github.com/microsoft/unilm/tree/master/beit # https://github.com/rwightman/pytorch-image-models/tree/master/timm # https://github.com/facebookresearch/deit/ # https://github.com/facebookresearch/dino # --------------------------------------------------------' import argparse import datetime import numpy as np import time import torch import torch.backends.cudnn as cudnn import json import os from pathlib import Path from timm.models import create_model from optim_factory import create_optimizer from datasets import build_vqkd_dataset from engine_for_vqkd import evaluate, train_one_epoch, calculate_codebook_usage from utils import NativeScalerWithGradNormCount as NativeScaler import utils import modeling_vqkd def get_args(): parser = argparse.ArgumentParser('BEiT pre-training script', add_help=False) parser.add_argument('--batch_size', default=64, type=int) parser.add_argument('--epochs', default=100, type=int) parser.add_argument('--save_ckpt_freq', default=20, type=int) # Model parameters parser.add_argument('--model', default='vqkd_encoder_base_decoder_3x768x12_clip', type=str, metavar='MODEL', help='Name of model to train') parser.add_argument('--rec_loss_type', default='cosine', type=str, metavar='MODEL', help='type of loss to calculate reconstruction distance') parser.add_argument('--codebook_n_emd', default=8192, type=int, metavar='MODEL', help='number of codebook') parser.add_argument('--codebook_emd_dim', default=32, type=int, metavar='MODEL', help='number of codebook') parser.add_argument('--ema_decay', default=0.99, type=float, metavar='MODEL', help='ema decay for quantizer') parser.add_argument('--quantize_kmeans_init', action='store_true', help='enable kmeans_init for quantizer') parser.add_argument('--process_type', default='default', type=str, choices=['default', 'dall-e', 'imagenet_norm'], help='Image process type (default, dall-e)') parser.add_argument('--input_size', default=224, type=int, help='images input size for backbone') # regress feature parser.add_argument('--teacher_model_type', default='clip', type=str, help='teacher_model_type during training') parser.add_argument('--teacher_input_size', default=224, type=int, help='teacher_input_size for clip-large p14') # Optimizer parameters parser.add_argument('--opt', default='adamw', type=str, metavar='OPTIMIZER', help='Optimizer (default: "adamw"') parser.add_argument('--opt_eps', default=1e-8, type=float, metavar='EPSILON', help='Optimizer Epsilon (default: 1e-8)') parser.add_argument('--opt_betas', default=None, type=float, nargs='+', metavar='BETA', help='Optimizer Betas (default: None, use opt default)') parser.add_argument('--clip_grad', type=float, default=None, metavar='NORM', help='Clip gradient norm (default: None, no clipping)') parser.add_argument('--weight_decay', type=float, default=1e-4, help='weight decay (default: 1e-4)') parser.add_argument('--weight_decay_end', type=float, default=None, help="""Final value of the weight decay. We use a cosine schedule for WD. (Set the same value with args.weight_decay to keep weight decay no change)""") parser.add_argument('--lr', type=float, default=5e-5, metavar='LR', help='learning rate (default: 5e-5)') parser.add_argument('--warmup_lr', type=float, default=1e-6, metavar='LR', help='warmup learning rate (default: 1e-6)') parser.add_argument('--min_lr', type=float, default=1e-5, metavar='LR', help='lower lr bound for cyclic schedulers that hit 0 (1e-5)') parser.add_argument('--warmup_epochs', type=int, default=5, metavar='N', help='epochs to warmup LR, if scheduler supports') parser.add_argument('--warmup_steps', type=int, default=-1, metavar='N', help='epochs to warmup LR, if scheduler supports') # Augmentation parameters parser.add_argument('--color_jitter', type=float, default=0., metavar='PCT', help='Color jitter factor (default: 0.)') parser.add_argument('--train_interpolation', type=str, default='bicubic', help='Training interpolation (random, bilinear, bicubic, lanczos default: "bicubic")') parser.add_argument('--min_crop_scale', type=float, default=0.08, metavar='PCT', help='min_crop_scale (default: 0.08)') # Dataset parameters parser.add_argument('--data_path', default='/datasets01/imagenet_full_size/061417/', type=str, help='dataset path') parser.add_argument('--eval_data_path', default='', type=str, help='dataset path') parser.add_argument('--data_set', default='image_folder', type=str, help='dataset path') parser.add_argument('--imagenet_default_mean_and_std', default=False, action='store_true') parser.add_argument('--output_dir', default='', help='path where to save, empty for no saving') parser.add_argument('--log_dir', default=None, 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('--auto_resume', action='store_true') parser.add_argument('--no_auto_resume', action='store_false', dest='auto_resume') parser.set_defaults(auto_resume=True) parser.add_argument('--dist_eval', action='store_true', default=True, help='Enabling distributed evaluation') parser.add_argument('--disable_eval', action='store_true', default=False) parser.add_argument('--eval', action='store_true', default=False, help="Perform evaluation only") parser.add_argument('--calculate_codebook_usage', action='store_true', default=False) parser.add_argument('--start_epoch', default=0, type=int, metavar='N', help='start epoch') 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', help='') 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.parse_args() def get_model(args, **kwargs): model = create_model( args.model, pretrained=False, as_tokenzer=False, n_code=args.codebook_n_emd, code_dim=args.codebook_emd_dim, img_size=args.input_size, rec_loss_type=args.rec_loss_type, teacher_model_type=args.teacher_model_type, teacher_input_size=args.teacher_input_size, decay=args.ema_decay, quantize_kmeans_init=args.quantize_kmeans_init, process_type=args.process_type ) return model def main(args): utils.init_distributed_mode(args) print(args) device = torch.device(args.device) # fix the seed for reproducibility seed = args.seed + utils.get_rank() torch.manual_seed(seed) np.random.seed(seed) # random.seed(seed) cudnn.benchmark = True model = get_model(args) # get dataset dataset_train = build_vqkd_dataset(is_train=True, args=args) if args.disable_eval: dataset_val = None else: dataset_val = build_vqkd_dataset(is_train=False, args=args) if True: # args.distributed: num_tasks = utils.get_world_size() global_rank = utils.get_rank() sampler_rank = global_rank num_training_steps_per_epoch = len(dataset_train) // args.batch_size // num_tasks sampler_train = torch.utils.data.DistributedSampler( dataset_train, num_replicas=num_tasks, rank=sampler_rank, shuffle=True ) print("Sampler_train = %s" % str(sampler_train)) if args.dist_eval: if len(dataset_val) % num_tasks != 0: print('Warning: Enabling distributed evaluation with an eval dataset not divisible by process number. ' 'This will slightly alter validation results as extra duplicate entries are added to achieve ' 'equal num of samples per-process.') sampler_val = torch.utils.data.DistributedSampler( dataset_val, num_replicas=num_tasks, rank=global_rank, shuffle=False) else: sampler_val = torch.utils.data.SequentialSampler(dataset_val) else: sampler_train = torch.utils.data.RandomSampler(dataset_train) sampler_val = torch.utils.data.SequentialSampler(dataset_val) if global_rank == 0 and args.log_dir is not None: os.makedirs(args.log_dir, exist_ok=True) log_writer = utils.TensorboardLogger(log_dir=args.log_dir) else: log_writer = None 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, ) if dataset_val is not None: data_loader_val = torch.utils.data.DataLoader( dataset_val, sampler=sampler_val, batch_size=int(1.5 * args.batch_size), num_workers=args.num_workers, pin_memory=args.pin_mem, drop_last=False ) else: data_loader_val = None model.to(device) model_without_ddp = model if not args.eval: print("Model = %s" % str(model_without_ddp)) for part in ['encoder', 'decoder']: model_part = eval(f"model.{part}") n_learnable_parameters = sum(p.numel() for p in model_part.parameters() if p.requires_grad) n_fix_parameters = sum(p.numel() for p in model_part.parameters() if not p.requires_grad) print(f'number of learnable params in model.{part}: {n_learnable_parameters / 1e6} M') print(f'number of fixed params in model.{part}: {n_fix_parameters / 1e6} M') n_learnable_parameters = sum(p.numel() for p in model.parameters() if p.requires_grad) n_fix_parameters = sum(p.numel() for p in model.parameters() if not p.requires_grad) print(f'total number of learnable params: {n_learnable_parameters / 1e6} M') print(f'total number of fixed params in : {n_fix_parameters / 1e6} M') total_batch_size = args.batch_size * utils.get_world_size() args.lr = total_batch_size / 128 * args.lr print("LR = %.8f" % args.lr) print("Min LR = %.8f" % args.min_lr) print("Weigth Decay = %.8f" % args.weight_decay) print("Batch size = %d" % total_batch_size) print("Number of training steps = %d" % num_training_steps_per_epoch) print("Number of training examples per epoch = %d" % (total_batch_size * num_training_steps_per_epoch)) optimizer = create_optimizer(args, model_without_ddp) loss_scaler = NativeScaler() if args.distributed: model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.gpu], find_unused_parameters=True) model_without_ddp = model.module print("Use step level LR & WD scheduler!") lr_schedule_values = utils.cosine_scheduler( args.lr, args.min_lr, args.epochs, num_training_steps_per_epoch, warmup_epochs=args.warmup_epochs, warmup_steps=args.warmup_steps, ) utils.auto_load_model( args=args, model=model, model_without_ddp=model_without_ddp, optimizer=optimizer, loss_scaler=loss_scaler) if args.eval: test_stats = evaluate(data_loader_val, model, device, log_writer, 0, args=args) exit(0) if args.calculate_codebook_usage: test_stats = calculate_codebook_usage(data_loader_val, model, device, log_writer, 0, args=args) exit(0) print(f"Start training for {args.epochs} epochs") start_time = time.time() for epoch in range(args.start_epoch, args.epochs): if args.distributed: data_loader_train.sampler.set_epoch(epoch) if log_writer is not None: log_writer.set_step(epoch * num_training_steps_per_epoch) train_stats = train_one_epoch( model, data_loader_train, optimizer, device, epoch, loss_scaler, args.clip_grad, log_writer=log_writer, start_steps=epoch * num_training_steps_per_epoch, lr_schedule_values=lr_schedule_values, args=args ) if args.output_dir: # if (epoch + 1) % args.save_ckpt_freq == 0 or epoch + 1 == args.epochs: utils.save_model( args=args, model=model, model_without_ddp=model_without_ddp, optimizer=optimizer, loss_scaler=loss_scaler, epoch=epoch, save_ckpt_freq=args.save_ckpt_freq) if data_loader_val is not None: test_stats = evaluate(data_loader_val, model, device, log_writer, epoch, args=args) print(f"Validation loss of the network on the {len(dataset_val)} test images: {test_stats['loss']:.4f}") if log_writer is not None: log_writer.update(**test_stats, head="val/loss") 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_learnable_parameters} else: log_stats = {**{f'train_{k}': v for k, v in train_stats.items()}, 'epoch': epoch, 'n_parameters': n_learnable_parameters} if args.output_dir and utils.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__': opts = get_args() if opts.output_dir: Path(opts.output_dir).mkdir(parents=True, exist_ok=True) main(opts)