From d6bffe1ed2518ef0bdb9763628d0b1c6bd5a4d5b Mon Sep 17 00:00:00 2001 From: Junnan Li Date: Fri, 16 Jul 2021 09:35:18 +0800 Subject: [PATCH] Delete Pretrain_resnet.py --- Pretrain_resnet.py | 190 --------------------------------------------- 1 file changed, 190 deletions(-) delete mode 100644 Pretrain_resnet.py diff --git a/Pretrain_resnet.py b/Pretrain_resnet.py deleted file mode 100644 index b4a8fad..0000000 --- a/Pretrain_resnet.py +++ /dev/null @@ -1,190 +0,0 @@ -import argparse -import os -import ruamel_yaml as yaml -import numpy as np -import random -import time -import datetime -import json -from pathlib import Path - -import torch -import torch.nn as nn -import torch.nn.functional as F -from torch.utils.data import DataLoader -import torch.backends.cudnn as cudnn -import torch.distributed as dist - -from models.model_pretrain_resnet import ALBEF -from models.tokenization_bert import BertTokenizer - -import utils -from dataset import create_dataset, create_sampler, create_loader -from scheduler import create_scheduler -from optim import create_optimizer - - -def train(model, data_loader, optimizer, tokenizer, epoch, warmup_steps, device, scheduler, config): - # train - model.train() - - metric_logger = utils.MetricLogger(delimiter=" ") - metric_logger.add_meter('lr', utils.SmoothedValue(window_size=50, fmt='{value:.6f}')) - metric_logger.add_meter('loss_mlm', utils.SmoothedValue(window_size=50, fmt='{value:.4f}')) - metric_logger.add_meter('loss_ita', utils.SmoothedValue(window_size=50, fmt='{value:.4f}')) - metric_logger.add_meter('loss_itm', utils.SmoothedValue(window_size=50, fmt='{value:.4f}')) - - header = 'Train Epoch: [{}]'.format(epoch) - print_freq = 50 - step_size = 100 - warmup_iterations = warmup_steps*step_size - - if args.distributed: - data_loader.sampler.set_epoch(epoch) - - for i, (image, text) in enumerate(metric_logger.log_every(data_loader, print_freq, header)): - - optimizer.zero_grad() - - image = image.to(device,non_blocking=True) - - text_input = tokenizer(text, padding='longest', truncation=True, max_length=25, return_tensors="pt").to(device) - - if epoch>0: - alpha = config['alpha'] - else: - alpha = config['alpha']*min(1,i/len(data_loader)) - - loss_mlm, loss_ita, loss_itm = model(image, text_input, alpha = alpha) - - loss = loss_mlm + loss_ita + loss_itm - - loss.backward() - optimizer.step() - - metric_logger.update(loss_mlm=loss_mlm.item()) - metric_logger.update(loss_ita=loss_ita.item()) - metric_logger.update(loss_itm=loss_itm.item()) - metric_logger.update(lr=optimizer.param_groups[0]["lr"]) - - if epoch==0 and i%step_size==0 and i<=warmup_iterations: - scheduler.step(i//step_size) - - # gather the stats from all processes - metric_logger.synchronize_between_processes() - print("Averaged stats:", metric_logger.global_avg()) - return {k: "{:.3f}".format(meter.global_avg) for k, meter in metric_logger.meters.items()} - - -def main(args, config): - utils.init_distributed_mode(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 - - start_epoch = 0 - max_epoch = config['schedular']['epochs'] - warmup_steps = config['schedular']['warmup_epochs'] - - #### Dataset #### - print("Creating dataset") - datasets = [create_dataset('pretrain', config)] - - if args.distributed: - num_tasks = utils.get_world_size() - global_rank = utils.get_rank() - samplers = create_sampler(datasets, [True], num_tasks, global_rank) - else: - samplers = [None] - - data_loader = create_loader(datasets,samplers,batch_size=[config['batch_size']], num_workers=[4], is_trains=[True], collate_fns=[None])[0] - - tokenizer = BertTokenizer.from_pretrained(args.text_encoder) - - #### Model #### - print("Creating model") - model = ALBEF(config=config, text_encoder=args.text_encoder, tokenizer=tokenizer) - - model = model.to(device) - - arg_opt = utils.AttrDict(config['optimizer']) - optimizer = create_optimizer(arg_opt, model) - arg_sche = utils.AttrDict(config['schedular']) - lr_scheduler, _ = create_scheduler(arg_sche, optimizer) - - - if args.checkpoint: - checkpoint = torch.load(args.checkpoint, map_location='cpu') - state_dict = checkpoint['model'] - if args.resume: - optimizer.load_state_dict(checkpoint['optimizer']) - lr_scheduler.load_state_dict(checkpoint['lr_scheduler']) - start_epoch = checkpoint['epoch']+1 - model.load_state_dict(state_dict) - print('load checkpoint from %s'%args.checkpoint) - - model_without_ddp = model - if args.distributed: - model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.gpu]) - model_without_ddp = model.module - - print("Start training") - start_time = time.time() - - for epoch in range(start_epoch, max_epoch): - - if epoch>0: - lr_scheduler.step(epoch+warmup_steps) - - train_stats = train(model, data_loader, optimizer, tokenizer, epoch, warmup_steps, device, lr_scheduler, config) - if utils.is_main_process(): - log_stats = {**{f'train_{k}': v for k, v in train_stats.items()}, - 'epoch': epoch, - } - save_obj = { - 'model': model_without_ddp.state_dict(), - 'optimizer': optimizer.state_dict(), - 'lr_scheduler': lr_scheduler.state_dict(), - 'config': config, - 'epoch': epoch, - } - torch.save(save_obj, os.path.join(args.output_dir, 'checkpoint_%02d.pth'%epoch)) - - with open(os.path.join(args.output_dir, "log.txt"),"a") as f: - f.write(json.dumps(log_stats) + "\n") - - dist.barrier() - - 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__': - parser = argparse.ArgumentParser() - parser.add_argument('--config', default='./configs/Pretrain.yaml') - parser.add_argument('--checkpoint', default='') - parser.add_argument('--resume', default=False, type=bool) - parser.add_argument('--output_dir', default='Pretrain/resnet101') - parser.add_argument('--text_encoder', default='bert-base-uncased') - parser.add_argument('--device', default='cuda') - parser.add_argument('--seed', default=42, type=int) - parser.add_argument('--world_size', default=1, type=int, help='number of distributed processes') - parser.add_argument('--dist_url', default='env://', help='url used to set up distributed training') - parser.add_argument('--distributed', default=True, type=bool) - args = parser.parse_args() - - config = yaml.load(open(args.config, 'r'), Loader=yaml.Loader) - - Path(args.output_dir).mkdir(parents=True, exist_ok=True) - - yaml.dump(config, open(os.path.join(args.output_dir, 'config.yaml'), 'w')) - - main(args, config) \ No newline at end of file