train.py

import os
from torch.backends import cudnn
from utils.logger import setup_logger
from datasets import make_dataloader
from model import make_model
from solver import make_optimizer, WarmupMultiStepLR
from loss import make_loss
from processor import do_train
import random
import torch
import numpy as np
import os
import argparse
from timm.scheduler import create_scheduler
from config import cfg
from timm.data import Mixup
def set_seed(seed):
    torch.manual_seed(seed)
    torch.cuda.manual_seed(seed)
    torch.cuda.manual_seed_all(seed)
    np.random.seed(seed)
    random.seed(seed)
    torch.backends.cudnn.deterministic = True
    torch.backends.cudnn.benchmark = True

if __name__ == '__main__':

    parser = argparse.ArgumentParser(description="ReID Baseline Training")
    parser.add_argument(
        "--config_file", default="", help="path to config file", type=str
    )

    parser.add_argument("opts", help="Modify config options using the command-line", default=None,
                        nargs=argparse.REMAINDER)
    parser.add_argument('--sched', default='cosine', type=str, metavar='SCHEDULER',
                        help='LR scheduler (default: "cosine"')
    #  parser.add_argument('--lr', type=float, default=0.01, metavar='LR',
                        #  help='learning rate (default: 5e-4)')
    parser.add_argument('--lr-noise', type=float, nargs='+', default=None, metavar='pct, pct',
                        help='learning rate noise on/off epoch percentages')
    parser.add_argument('--lr-noise-pct', type=float, default=0.67, metavar='PERCENT',
                        help='learning rate noise limit percent (default: 0.67)')
    parser.add_argument('--lr-noise-std', type=float, default=1.0, metavar='STDDEV',
                        help='learning rate noise std-dev (default: 1.0)')
    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('--cooldown-epochs', type=int, default=10, metavar='N',
                        help='epochs to cooldown LR at min_lr, after cyclic schedule ends')
    parser.add_argument('--decay-rate', '--dr', type=float, default=0.1, metavar='RATE',
                        help='LR decay rate (default: 0.1)')
    parser.add_argument('--epochs', default=120, type=int)
    parser.add_argument("--local_rank", default=0, type=int)
    args = parser.parse_args()
    
    if args.config_file != "":
        cfg.merge_from_file(args.config_file)
    cfg.merge_from_list(args.opts)
    cfg.TEST.CROP_TEST = False
    cfg.freeze()

    set_seed(cfg.SOLVER.SEED)
    if cfg.MODEL.DIST_TRAIN:
        torch.cuda.set_device(args.local_rank)
    else:
        pass

    output_dir = cfg.OUTPUT_DIR
    if output_dir and not os.path.exists(output_dir):
        os.makedirs(output_dir)

    logger = setup_logger("reid_baseline", output_dir, if_train=True)
    logger.info("Saving model in the path :{}".format(cfg.OUTPUT_DIR))

    if args.config_file != "":
        logger.info("Loaded configuration file {}".format(args.config_file))
        with open(args.config_file, 'r') as cf:
            config_str = "\n" + cf.read()
    logger.info("Running with config:\n{}".format(cfg))

    if cfg.MODEL.DIST_TRAIN:
        torch.distributed.init_process_group(backend='nccl', init_method='env://')

    os.environ['CUDA_VISIBLE_DEVICES'] = cfg.MODEL.DEVICE_ID
    train_loader, val_loader, num_query, num_classes = make_dataloader(cfg)
    model = make_model(cfg, num_class=num_classes)

    loss_func, center_criterion = make_loss(cfg, num_classes=num_classes)

    optimizer, optimizer_center = make_optimizer(cfg, model, center_criterion)
    args.sched = cfg.SOLVER.WARMUP_METHOD
    args.epochs = cfg.SOLVER.MAX_EPOCHS
    args.warmup_epochs = cfg.SOLVER.WARMUP_EPOCHS

    if args.sched == 'cosine':
        print('===========using cosine learning rate=======')
        scheduler, _ = create_scheduler(args, optimizer)
    else:
        print('===========using normal learning rate=======')
        scheduler = WarmupMultiStepLR(optimizer, cfg.SOLVER.STEPS, cfg.SOLVER.GAMMA,
                                      cfg.SOLVER.WARMUP_FACTOR,
                                      cfg.SOLVER.WARMUP_EPOCHS, cfg.SOLVER.WARMUP_METHOD)

    do_train(
        cfg,
        model,
        center_criterion,
        train_loader,
        val_loader,
        optimizer,
        optimizer_center,
        scheduler,  # modify for using self trained model
        loss_func,
        num_query, args.local_rank
    )
    print(cfg.OUTPUT_DIR)