linear_BE_data2vec.py

# --------------------------------------------------------
# BEIT: BERT Pre-Training of Image Transformers (https://arxiv.org/abs/2106.08254)
# Github source: https://github.com/microsoft/unilm/tree/master/beit
# Copyright (c) 2021 Microsoft
# Licensed under The MIT License [see LICENSE for details]
# By Hangbo Bao
# Based on timm, DINO and DeiT code bases
# https://github.com/rwightman/pytorch-image-models/tree/master/timm
# https://github.com/facebookresearch/deit
# https://github.com/facebookresearch/dino
# --------------------------------------------------------'
# Copyright (c) Meta Platforms, Inc. and affiliates
import argparse
import datetime
import numpy as np
import time
import torch
import torch.nn as nn
import torch.backends.cudnn as cudnn
import json
import os

from pathlib import Path

from timm.data.mixup import Mixup
from timm.models import create_model
from timm.loss import LabelSmoothingCrossEntropy, SoftTargetCrossEntropy
from timm.utils import ModelEma
from models.data2vec.optim_factory import create_optimizer, get_parameter_groups, LayerDecayValueAssigner

from models.data2vec.engine_for_finetuning import train_one_epoch, evaluate
from models.data2vec.utils import NativeScalerWithGradNormCount as NativeScaler
import models.data2vec.utils as utils
from scipy import interpolate
import models.data2vec.modeling_finetune as modeling_finetune
from sklearn.metrics import average_precision_score
from datasets.BigEarthNet.bigearthnet_dataset_seco_lmdb_s2_uint8 import LMDBDataset, random_subset

from cvtorchvision import cvtransforms

def build_dataset_be(is_train, args):
    if args.bands == 'RGB':
        bands = ['B04', 'B03', 'B02']
        lmdb_train = 'train_RGB.lmdb'
        lmdb_val = 'val_RGB.lmdb'
    else:
        bands = ['B01', 'B02', 'B03', 'B04', 'B05', 'B06', 'B07', 'B08', 'B8A', 'B09', 'B11', 'B12']
        lmdb_train = 'train_B12.lmdb'
        lmdb_val = 'val_B12.lmdb'

    ## change03 ##
    train_transforms = cvtransforms.Compose([
            cvtransforms.RandomResizedCrop(args.in_size,scale=(0.8,1.0)),
            cvtransforms.RandomHorizontalFlip(),
            cvtransforms.ToTensor()])

    val_transforms = cvtransforms.Compose([
            cvtransforms.Resize(256),
            cvtransforms.CenterCrop(args.in_size),
            cvtransforms.ToTensor(),
            ])


    train_dataset = LMDBDataset(
        lmdb_file=os.path.join(args.data_path, lmdb_train),
        transform=train_transforms,
        is_slurm_job=False            
    )

    val_dataset = LMDBDataset(
        lmdb_file=os.path.join(args.data_path, lmdb_val),
        transform=val_transforms,
        is_slurm_job=False            
    )       
    
    if args.train_frac is not None and args.train_frac<1:
        train_dataset = random_subset(train_dataset,args.train_frac,seed=42)    
    
    return train_dataset, val_dataset, 19



def get_args():
    parser = argparse.ArgumentParser('BEiT fine-tuning and evaluation script for image classification', add_help=False)
    parser.add_argument('--batch_size', default=64, type=int)
    parser.add_argument('--epochs', default=30, type=int)
    parser.add_argument('--update_freq', default=1, type=int)
    parser.add_argument('--save_ckpt_freq', default=5, type=int)

    # Model parameters
    parser.add_argument('--model', default='deit_base_patch16_224', type=str, metavar='MODEL',
                        help='Name of model to train')
    parser.add_argument('--rel_pos_bias', action='store_true')
    parser.add_argument('--disable_rel_pos_bias', action='store_false', dest='rel_pos_bias')
    parser.set_defaults(rel_pos_bias=True)
    parser.add_argument('--abs_pos_emb', action='store_true')
    parser.set_defaults(abs_pos_emb=False)
    parser.add_argument('--layer_scale_init_value', default=0.1, type=float, 
                        help="0.1 for base, 1e-5 for large. set 0 to disable layer scale")

    parser.add_argument('--input_size', default=224, type=int,
                        help='images input size')

    parser.add_argument('--drop', type=float, default=0.0, metavar='PCT',
                        help='Dropout rate (default: 0.)')
    parser.add_argument('--attn_drop_rate', type=float, default=0.0, metavar='PCT',
                        help='Attention dropout rate (default: 0.)')
    parser.add_argument('--drop_path', type=float, default=0.1, metavar='PCT',
                        help='Drop path rate (default: 0.1)')

    parser.add_argument('--disable_eval_during_finetuning', action='store_true', default=False)

    parser.add_argument('--model_ema', action='store_true', default=False)
    parser.add_argument('--model_ema_decay', type=float, default=0.9999, help='')
    parser.add_argument('--model_ema_force_cpu', action='store_true', default=False, help='')

    # Optimizer parameters
    parser.add_argument('--opt', default='sgd', 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('--momentum', type=float, default=0.9, metavar='M',
                        help='SGD momentum (default: 0.9)')
    parser.add_argument('--weight_decay', type=float, default=0.05,
                        help='weight decay (default: 0.05)')
    parser.add_argument('--weight_decay_end', type=float, default=None, help="""Final value of the
        weight decay. We use a cosine schedule for WD and using a larger decay by
        the end of training improves performance for ViTs.""")

    parser.add_argument('--lr', type=float, default=5e-4, metavar='LR',
                        help='learning rate (default: 5e-4)')
    parser.add_argument('--layer_decay', type=float, default=0.9)

    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-6, 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='num of steps to warmup LR, will overload warmup_epochs if set > 0')

    # Augmentation parameters
    parser.add_argument('--color_jitter', type=float, default=0.4, metavar='PCT',
                        help='Color jitter factor (default: 0.4)')
    parser.add_argument('--aa', type=str, default='rand-m9-mstd0.5-inc1', metavar='NAME',
                        help='Use AutoAugment policy. "v0" or "original". " + "(default: rand-m9-mstd0.5-inc1)'),
    parser.add_argument('--smoothing', type=float, default=0.1,
                        help='Label smoothing (default: 0.1)')
    parser.add_argument('--train_interpolation', type=str, default='bicubic',
                        help='Training interpolation (random, bilinear, bicubic default: "bicubic")')

    # Evaluation parameters
    parser.add_argument('--crop_pct', type=float, default=None)

    # * Random Erase params
    parser.add_argument('--reprob', type=float, default=0.25, metavar='PCT',
                        help='Random erase prob (default: 0.25)')
    parser.add_argument('--remode', type=str, default='pixel',
                        help='Random erase mode (default: "pixel")')
    parser.add_argument('--recount', type=int, default=1,
                        help='Random erase count (default: 1)')
    parser.add_argument('--resplit', action='store_true', default=False,
                        help='Do not random erase first (clean) augmentation split')

    # * Mixup params
    parser.add_argument('--mixup', type=float, default=0,
                        help='mixup alpha, mixup enabled if > 0.')
    parser.add_argument('--cutmix', type=float, default=0,
                        help='cutmix alpha, cutmix enabled if > 0.')
    parser.add_argument('--cutmix_minmax', type=float, nargs='+', default=None,
                        help='cutmix min/max ratio, overrides alpha and enables cutmix if set (default: None)')
    parser.add_argument('--mixup_prob', type=float, default=1.0,
                        help='Probability of performing mixup or cutmix when either/both is enabled')
    parser.add_argument('--mixup_switch_prob', type=float, default=0.5,
                        help='Probability of switching to cutmix when both mixup and cutmix enabled')
    parser.add_argument('--mixup_mode', type=str, default='batch',
                        help='How to apply mixup/cutmix params. Per "batch", "pair", or "elem"')

    # * Finetuning params
    parser.add_argument('--finetune', default='',
                        help='finetune from checkpoint')
    parser.add_argument('--model_key', default='model|module', type=str)
    parser.add_argument('--model_prefix', default='', type=str)
    parser.add_argument('--init_scale', default=0.001, type=float)
    parser.add_argument('--use_mean_pooling', action='store_true')
    parser.set_defaults(use_mean_pooling=True)
    parser.add_argument('--use_cls', action='store_false', dest='use_mean_pooling')
    parser.add_argument('--disable_weight_decay_on_rel_pos_bias', action='store_true', default=False)
    parser.add_argument('--target_layer', default=-1, type=int, help="target output layer (0-based)")
    parser.add_argument('--remove_final_norm', action='store_true', dest='remove_final_norm')
    parser.add_argument('--reinit_final_norm', action='store_true', dest='reinit_final_norm')
    parser.add_argument('--learn_layer_weights', action='store_true', dest='learn_layer_weights')  # supersede `target_layer`
    parser.add_argument('--layernorm_before_combine', action='store_true', dest='layernorm_before_combine')

    # 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=None, type=str,
                        help='dataset path for evaluation')
    parser.add_argument('--nb_classes', default=0, type=int,
                        help='number of the classification types')
    parser.add_argument('--linear_classifier', action='store_true',
                        help='linear classifier')
    parser.add_argument('--imagenet_default_mean_and_std', default=False, action='store_true')

    parser.add_argument('--data_set', default='IMNET', choices=['CIFAR', 'IMNET', 'image_folder'],
                        type=str, help='ImageNet dataset path')
    parser.add_argument('--data_set_filter_file', type=str, default=None, help="path to filter to filter dataset")
    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=42, 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('--save_ckpt', action='store_true')
    parser.add_argument('--no_save_ckpt', action='store_false', dest='save_ckpt')
    parser.set_defaults(save_ckpt=True)

    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')
    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')

    parser.add_argument('--enable_deepspeed', action='store_true', default=False)
    parser.add_argument('--bands', default='all',
                        help='which bands to use')

    parser.add_argument(
        "--num_mask_patches",
        default=0,
        type=int,
        help="number of the visual tokens/patches need be masked",
    )
    parser.add_argument("--max_mask_patches_per_block", type=int, default=None)
    parser.add_argument("--min_mask_patches_per_block", type=int, default=16)
    
    parser.add_argument('--train_frac', type=float, default=0.1,
                        help='Training set size')
    parser.add_argument('--in_size', type=int, default=224,
                        help='Training set size')

    known_args, _ = parser.parse_known_args()

    if known_args.enable_deepspeed:
        try:
            print("why")
            import deepspeed
            print("Imported deepspeed")
            from deepspeed import DeepSpeedConfig
            print("Imported config")
            parser = deepspeed.add_config_arguments(parser)
            print("Created parser")
            ds_init = deepspeed.initialize
            print("Inited deepspeed")
        except:
            print("Please 'pip install deepspeed==0.4.0'")
            exit(0)
    else:
        ds_init = None

    return parser.parse_args(), ds_init


def main(args, ds_init):
    utils.init_distributed_mode(args)

    if ds_init is not None:
        utils.create_ds_config(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

    dataset_train, dataset_val, args.nb_classes = build_dataset_be(is_train=True, args=args)
    #if args.disable_eval_during_finetuning:
    #    dataset_val = None
    #else:
    #    dataset_val, _ = build_dataset(is_train=False, args=args)
        
        

    if True:  # args.distributed:
        num_tasks = utils.get_world_size()
        global_rank = utils.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))
        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

    mixup_fn = None
    mixup_active = args.mixup > 0 or args.cutmix > 0. or args.cutmix_minmax is not None
    if mixup_active:
        print("Mixup is activated!")
        mixup_fn = Mixup(
            mixup_alpha=args.mixup, cutmix_alpha=args.cutmix, cutmix_minmax=args.cutmix_minmax,
            prob=args.mixup_prob, switch_prob=args.mixup_switch_prob, mode=args.mixup_mode,
            label_smoothing=args.smoothing, num_classes=args.nb_classes)

    model = create_model(
        args.model,
        pretrained=False,
        num_classes=args.nb_classes,
        drop_rate=args.drop,
        drop_path_rate=args.drop_path,
        attn_drop_rate=args.attn_drop_rate,
        drop_block_rate=None,
        use_mean_pooling=args.use_mean_pooling,
        init_scale=args.init_scale,
        use_rel_pos_bias=False,
        use_shared_rel_pos_bias=args.rel_pos_bias,
        use_abs_pos_emb=args.abs_pos_emb,
        init_values=args.layer_scale_init_value,
        linear_classifier=args.linear_classifier,
        has_masking=args.num_mask_patches > 0,
        learn_layer_weights=args.learn_layer_weights,
        layernorm_before_combine=args.layernorm_before_combine,
    )

    patch_size = model.patch_embed.patch_size
    print("Patch size = %s" % str(patch_size))
    args.window_size = (args.input_size // patch_size[0], args.input_size // patch_size[1])
    args.patch_size = patch_size

    masked_position_generator = None
    if args.num_mask_patches > 0:
        from masking_generator import MaskingGenerator
        masked_position_generator = MaskingGenerator(
            args.window_size, num_masking_patches=args.num_mask_patches,
            max_num_patches=args.max_mask_patches_per_block,
            min_num_patches=args.min_mask_patches_per_block,
        )

    if args.finetune:
        if args.finetune.startswith('https'):
            checkpoint = torch.hub.load_state_dict_from_url(
                args.finetune, map_location='cpu', check_hash=True)
        else:
            checkpoint = torch.load(args.finetune, map_location='cpu')

        print("Load ckpt from %s" % args.finetune)
        checkpoint_model = None
        for model_key in args.model_key.split('|'):
            if model_key in checkpoint:
                checkpoint_model = checkpoint[model_key]
                print("Load state_dict by model_key = %s" % model_key)
                break
        if checkpoint_model is None:
            checkpoint_model = checkpoint
        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]
        if args.reinit_final_norm:
            for k in ['norm.weight', 'norm.bias', 'fc_norm.weight', 'fc_norm.bias']:
                if k in checkpoint_model:
                    print(f"Removing key {k} from pretrained checkpoint")
                    del checkpoint_model[k]

        if model.use_rel_pos_bias and "rel_pos_bias.relative_position_bias_table" in checkpoint_model:
            print("Expand the shared relative position embedding to each transformer block. ")
            num_layers = model.get_num_layers()
            rel_pos_bias = checkpoint_model["rel_pos_bias.relative_position_bias_table"]
            for i in range(num_layers):
                checkpoint_model["blocks.%d.attn.relative_position_bias_table" % i] = rel_pos_bias.clone()

            checkpoint_model.pop("rel_pos_bias.relative_position_bias_table")

        all_keys = list(checkpoint_model.keys())
        for key in all_keys:
            if "relative_position_index" in key:
                checkpoint_model.pop(key)

            if "relative_position_bias_table" in key:
                rel_pos_bias = checkpoint_model[key]
                src_num_pos, num_attn_heads = rel_pos_bias.size()
                dst_num_pos, _ = model.state_dict()[key].size()
                dst_patch_shape = model.patch_embed.patch_shape
                if dst_patch_shape[0] != dst_patch_shape[1]:
                    raise NotImplementedError()
                num_extra_tokens = dst_num_pos - (dst_patch_shape[0] * 2 - 1) * (dst_patch_shape[1] * 2 - 1)
                src_size = int((src_num_pos - num_extra_tokens) ** 0.5)
                dst_size = int((dst_num_pos - num_extra_tokens) ** 0.5)
                if src_size != dst_size:
                    print("Position interpolate for %s from %dx%d to %dx%d" % (
                        key, src_size, src_size, dst_size, dst_size))
                    extra_tokens = rel_pos_bias[-num_extra_tokens:, :]
                    rel_pos_bias = rel_pos_bias[:-num_extra_tokens, :]

                    def geometric_progression(a, r, n):
                        return a * (1.0 - r ** n) / (1.0 - r)

                    left, right = 1.01, 1.5
                    while right - left > 1e-6:
                        q = (left + right) / 2.0
                        gp = geometric_progression(1, q, src_size // 2)
                        if gp > dst_size // 2:
                            right = q
                        else:
                            left = q

                    # if q > 1.090307:
                    #     q = 1.090307

                    dis = []
                    cur = 1
                    for i in range(src_size // 2):
                        dis.append(cur)
                        cur += q ** (i + 1)

                    r_ids = [-_ for _ in reversed(dis)]

                    x = r_ids + [0] + dis
                    y = r_ids + [0] + dis

                    t = dst_size // 2.0
                    dx = np.arange(-t, t + 0.1, 1.0)
                    dy = np.arange(-t, t + 0.1, 1.0)

                    print("Original positions = %s" % str(x))
                    print("Target positions = %s" % str(dx))

                    all_rel_pos_bias = []

                    for i in range(num_attn_heads):
                        z = rel_pos_bias[:, i].view(src_size, src_size).float().numpy()
                        f = interpolate.interp2d(x, y, z, kind='cubic')
                        all_rel_pos_bias.append(
                            torch.Tensor(f(dx, dy)).contiguous().view(-1, 1).to(rel_pos_bias.device))

                    rel_pos_bias = torch.cat(all_rel_pos_bias, dim=-1)

                    new_rel_pos_bias = torch.cat((rel_pos_bias, extra_tokens), dim=0)
                    checkpoint_model[key] = new_rel_pos_bias

        # interpolate position embedding
        if 'pos_embed' in checkpoint_model:
            pos_embed_checkpoint = checkpoint_model['pos_embed']
            embedding_size = pos_embed_checkpoint.shape[-1]
            num_patches = model.patch_embed.num_patches
            num_extra_tokens = model.pos_embed.shape[-2] - num_patches
            # height (== width) for the checkpoint position embedding
            orig_size = int((pos_embed_checkpoint.shape[-2] - num_extra_tokens) ** 0.5)
            # height (== width) for the new position embedding
            new_size = int(num_patches ** 0.5)
            # class_token and dist_token are kept unchanged
            if orig_size != new_size:
                print("Position interpolate from %dx%d to %dx%d" % (orig_size, orig_size, new_size, new_size))
                extra_tokens = pos_embed_checkpoint[:, :num_extra_tokens]
                # only the position tokens are interpolated
                pos_tokens = pos_embed_checkpoint[:, num_extra_tokens:]
                pos_tokens = pos_tokens.reshape(-1, orig_size, orig_size, embedding_size).permute(0, 3, 1, 2)
                pos_tokens = torch.nn.functional.interpolate(
                    pos_tokens, size=(new_size, new_size), mode='bicubic', align_corners=False)
                pos_tokens = pos_tokens.permute(0, 2, 3, 1).flatten(1, 2)
                new_pos_embed = torch.cat((extra_tokens, pos_tokens), dim=1)
                checkpoint_model['pos_embed'] = new_pos_embed

        if not args.learn_layer_weights and args.target_layer != -1:
            print(f"model target layer is {args.target_layer}")
            model.blocks = model.blocks[:args.target_layer+1]

        if args.remove_final_norm:
            print(f"removing final norm by replacing it with Identity")
            model.norm = None if model.norm is None else nn.Identity()
            model.fc_norm = None if model.fc_norm is None else nn.Identity()

        if args.linear_classifier:
            frozen_params = (
                set(n for n, _ in model.named_parameters())
                & set(checkpoint_model.keys())
            )
            for n, p in model.named_parameters():
                if n in frozen_params:
                    p.requires_grad_(False)
            param_names = [n for n, p in model.named_parameters() if p.requires_grad]
            print(f"Trainable weights: {param_names}")

        utils.load_state_dict(model, checkpoint_model, prefix=args.model_prefix)
        # model.load_state_dict(checkpoint_model, strict=False)

    model.to(device)

    model_ema = None
    if args.model_ema:
        # Important to create EMA model after cuda(), DP wrapper, and AMP but before SyncBN and DDP wrapper
        model_ema = ModelEma(
            model,
            decay=args.model_ema_decay,
            device='cpu' if args.model_ema_force_cpu else '',
            resume='')
        print("Using EMA with decay = %.8f" % args.model_ema_decay)

    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:', n_parameters)

    total_batch_size = args.batch_size * args.update_freq * utils.get_world_size()
    num_training_steps_per_epoch = len(dataset_train) // total_batch_size
    print("LR = %.8f" % args.lr)
    print("Batch size = %d" % total_batch_size)
    print("Update frequent = %d" % args.update_freq)
    print("Number of training examples = %d" % len(dataset_train))
    print("Number of training training per epoch = %d" % num_training_steps_per_epoch)

    num_layers = model_without_ddp.get_num_layers()
    if args.layer_decay < 1.0:
        assigner = LayerDecayValueAssigner(list(args.layer_decay ** (num_layers + 1 - i) for i in range(num_layers + 2)))
    else:
        assigner = None

    if assigner is not None:
        print("Assigned values = %s" % str(assigner.values))

    skip_weight_decay_list = model.no_weight_decay()
    if args.disable_weight_decay_on_rel_pos_bias:
        for i in range(num_layers):
            skip_weight_decay_list.add("blocks.%d.attn.relative_position_bias_table" % i)

    if args.enable_deepspeed:
        loss_scaler = None
        optimizer_params = get_parameter_groups(
            model, args.weight_decay, skip_weight_decay_list,
            assigner.get_layer_id if assigner is not None else None,
            assigner.get_scale if assigner is not None else None)
        model, optimizer, _, _ = ds_init(
            args=args, model=model, model_parameters=optimizer_params, dist_init_required=not args.distributed,
        )

        print("model.gradient_accumulation_steps() = %d" % model.gradient_accumulation_steps())
        assert model.gradient_accumulation_steps() == args.update_freq
    else:
        if args.distributed:
            model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.gpu], find_unused_parameters=True)
            model_without_ddp = model.module

        optimizer = create_optimizer(
            args, model_without_ddp, skip_list=skip_weight_decay_list,
            get_num_layer=assigner.get_layer_id if assigner is not None else None, 
            get_layer_scale=assigner.get_scale if assigner is not None else None)
        loss_scaler = NativeScaler()

    print("Use step level LR 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,
    )
    if args.weight_decay_end is None:
        args.weight_decay_end = args.weight_decay
    wd_schedule_values = utils.cosine_scheduler(
        args.weight_decay, args.weight_decay_end, args.epochs, num_training_steps_per_epoch)
    print("Max WD = %.7f, Min WD = %.7f" % (max(wd_schedule_values), min(wd_schedule_values)))

    #if mixup_fn is not None:
        # smoothing is handled with mixup label transform
    #    criterion = SoftTargetCrossEntropy()
    #elif args.smoothing > 0.:
    #    criterion = LabelSmoothingCrossEntropy(smoothing=args.smoothing)
    #else:
    
    criterion = torch.nn.MultiLabelSoftMarginLoss()

    print("criterion = %s" % str(criterion))

    utils.auto_load_model(
        args=args, model=model, model_without_ddp=model_without_ddp,
        optimizer=optimizer, loss_scaler=loss_scaler, model_ema=model_ema)

    if args.eval:
        test_stats = evaluate(data_loader_val, model, device, metric='map', padd=True)
        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)
        if log_writer is not None:
            log_writer.set_step(epoch * num_training_steps_per_epoch * args.update_freq)
        train_stats = train_one_epoch(
            model, criterion, data_loader_train, optimizer,
            device, epoch, loss_scaler, args.clip_grad, model_ema, mixup_fn,
            log_writer=log_writer, start_steps=epoch * num_training_steps_per_epoch,
            lr_schedule_values=lr_schedule_values, wd_schedule_values=wd_schedule_values,
            num_training_steps_per_epoch=num_training_steps_per_epoch, update_freq=args.update_freq,
            masked_position_generator=masked_position_generator,
            metric='map', padd=True
        )
        if args.output_dir and args.save_ckpt:
            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, model_ema=model_ema)
        if data_loader_val is not None and epoch % 5 == 0:
            test_stats = evaluate(data_loader_val, model, device, 'map', padd=True)
            print(f"Accuracy of the network on the {len(dataset_val)} test images: {test_stats['acc1']:.1f}%")
            if max_accuracy < test_stats["acc1"]:
                max_accuracy = test_stats["acc1"]
                if args.output_dir and args.save_ckpt:
                    utils.save_model(
                        args=args, model=model, model_without_ddp=model_without_ddp, optimizer=optimizer,
                        loss_scaler=loss_scaler, epoch="best", model_ema=model_ema)

            print(f'Max accuracy: {max_accuracy:.2f}%')
            if log_writer is not None:
                log_writer.update(test_acc1=test_stats['acc1'], head="perf", step=epoch)
                log_writer.update(test_acc5=test_stats['acc5'], head="perf", step=epoch)
                log_writer.update(test_loss=test_stats['loss'], head="perf", step=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}
        else:
            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 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, ds_init = get_args()
    if opts.output_dir:
        Path(opts.output_dir).mkdir(parents=True, exist_ok=True)
    main(opts, ds_init)