eval_finetune.py

"""
Mostly copy-pasted and adapted from the MAE[*] code: https://github.com/facebookresearch/mae
[*] Masked Autoencoders Are Scalable Vision Learners, CVPR'22
References:
timm: https://github.com/rwightman/pytorch-image-models/tree/master/timm
DeiT: https://github.com/facebookresearch/deit
ELECTRA https://github.com/google-research/electra
BEiT: https://github.com/microsoft/unilm/tree/master/beit
"""

import argparse
import datetime
import json
import numpy as np
import os
import time
from pathlib import Path
from typing import Iterable, Optional
import PIL
import math
import sys

import torch
import torch.nn as nn
import torch.backends.cudnn as cudnn
from torch.utils.tensorboard import SummaryWriter

import timm

from timm.models.layers import trunc_normal_
#from timm.data import Mixup
from timm.data.mixup import Mixup
from timm.loss import LabelSmoothingCrossEntropy, SoftTargetCrossEntropy
from timm.data import create_transform
from timm.data.constants import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
from timm.utils import accuracy
from torchvision import datasets, transforms

import util.misc as misc
import util.data_utils as data_utils
from util.model_utils import load_pretrained_encoder
from util.misc import NativeScalerWithGradNormCount as NativeScaler

import models_vit


def get_args_parser():
    parser = argparse.ArgumentParser('Fine-tuning for image classification', add_help=False)
    parser.add_argument('--batch_size', default=64, type=int,
                        help='Batch size per GPU (effective batch size is batch_size * accum_iter * # gpus')
    parser.add_argument('--epochs', default=100, 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_base_patch16', type=str, metavar='MODEL',
                        help='Name of model to train')

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

    parser.add_argument('--drop_path', type=float, default=0.2, metavar='PCT',
                        help='Drop path rate (ibot default: 0.2)')

    # Optimizer parameters
    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=0.05,
                        help='weight decay (default: 0.05)')

    parser.add_argument('--lr', type=float, default=None, metavar='LR',
                        help='learning rate (absolute lr)')
    parser.add_argument('--blr', type=float, default=1e-3, metavar='LR',
                        help='base learning rate: absolute_lr = base_lr * total_batch_size / 256')
    parser.add_argument('--layer_decay', type=float, default=0.65,
                        help='layer-wise lr decay from ELECTRA/BEiT (ibot default: 0.65)')

    parser.add_argument('--min_lr', type=float, default=1e-6, metavar='LR',
                        help='lower lr bound for cyclic schedulers that hit 0')

    parser.add_argument('--warmup_epochs', type=int, default=5, metavar='N',
                        help='epochs to warmup LR')

    # Augmentation parameters
    parser.add_argument('--color_jitter', type=float, default=None, metavar='PCT',
                        help='Color jitter factor (enabled only when not using Auto/RandAug)')
    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)')

    # * 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('--global_pool', action='store_true')
    parser.add_argument('--use_fc_norm', action='store_true')

    # Dataset parameters
    parser.add_argument('--data_path', default='/datasets_local/ImageNet/', type=str,
                        help='dataset path')
    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('--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('--subset', default=-1, type=int, help='The number of images per class that they would be use for '
                        'training (default -1). If -1, then all the availabe images are used.')

    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=8, 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')
    #Only for Hades
    parser.add_argument('--gpu', default=0, type=int, help='which gpu')
    parser.add_argument('--backend', default='nccl', type=str, help='Specify backend nccl or gloo')

    parser.add_argument("--use_teacher", action='store_true')
    parser.add_argument('--resume_if_exists', action='store_true')

    return parser


def adjust_learning_rate(optimizer, epoch, args):
    """Decay the learning rate with half-cycle cosine after warmup"""
    if epoch < args.warmup_epochs:
        lr = args.lr * epoch / args.warmup_epochs 
    else:
        lr = args.min_lr + (args.lr - args.min_lr) * 0.5 * \
            (1. + math.cos(math.pi * (epoch - args.warmup_epochs) / (args.epochs - args.warmup_epochs)))
    for param_group in optimizer.param_groups:
        if "lr_scale" in param_group:
            param_group["lr"] = lr * param_group["lr_scale"]
        else:
            param_group["lr"] = lr
    return lr


def build_dataset(is_train, args):
    transform = build_transform(is_train, args)

    root = os.path.join(args.data_path, 'train' if is_train else 'val')
    dataset = datasets.ImageFolder(root, transform=transform)

    print(dataset)

    return dataset


def build_transform(is_train, args):
    mean = IMAGENET_DEFAULT_MEAN
    std = IMAGENET_DEFAULT_STD
    # train transform
    if is_train:
        # this should always dispatch to transforms_imagenet_train
        transform = create_transform(
            input_size=args.input_size,
            is_training=True,
            color_jitter=args.color_jitter,
            auto_augment=args.aa,
            interpolation='bicubic',
            re_prob=args.reprob,
            re_mode=args.remode,
            re_count=args.recount,
            mean=mean,
            std=std,
        )
        return transform

    # eval transform
    t = []
    if args.input_size <= 224:
        crop_pct = 224 / 256
    else:
        crop_pct = 1.0
    size = int(args.input_size / crop_pct)
    t.append(
        transforms.Resize(size, interpolation=PIL.Image.BICUBIC),  # to maintain same ratio w.r.t. 224 images
    )
    t.append(transforms.CenterCrop(args.input_size))

    t.append(transforms.ToTensor())
    t.append(transforms.Normalize(mean, std))
    return transforms.Compose(t)


def param_groups_lrd(model, weight_decay=0.05, no_weight_decay_list=[], layer_decay=.75):
    """
    Parameter groups for layer-wise lr decay
    Following BEiT: https://github.com/microsoft/unilm/blob/master/beit/optim_factory.py#L58
    """
    param_group_names = {}
    param_groups = {}

    num_layers = len(model.blocks) + 1

    layer_scales = list(layer_decay ** (num_layers - i) for i in range(num_layers + 1))

    for n, p in model.named_parameters():
        if not p.requires_grad:
            continue

        # no decay: all 1D parameters and model specific ones
        if p.ndim == 1 or n in no_weight_decay_list:
            g_decay = "no_decay"
            this_decay = 0.
        else:
            g_decay = "decay"
            this_decay = weight_decay
            
        layer_id = get_layer_id_for_vit(n, num_layers)
        group_name = "layer_%d_%s" % (layer_id, g_decay)

        if group_name not in param_group_names:
            this_scale = layer_scales[layer_id]

            param_group_names[group_name] = {
                "lr_scale": this_scale,
                "weight_decay": this_decay,
                "params": [],
            }
            param_groups[group_name] = {
                "lr_scale": this_scale,
                "weight_decay": this_decay,
                "params": [],
            }

        param_group_names[group_name]["params"].append(n)
        param_groups[group_name]["params"].append(p)

    # print("parameter groups: \n%s" % json.dumps(param_group_names, indent=2))

    return list(param_groups.values())


def get_layer_id_for_vit(name, num_layers):
    """
    Assign a parameter with its layer id
    Following BEiT: https://github.com/microsoft/unilm/blob/master/beit/optim_factory.py#L33
    """
    if name in ['cls_token', 'pos_embed']:
        return 0
    elif name.startswith('patch_embed'):
        return 0
    elif name.startswith('blocks'):
        return int(name.split('.')[1]) + 1
    else:
        return num_layers


def train_one_epoch(model: torch.nn.Module, criterion: torch.nn.Module,
                    data_loader: Iterable, optimizer: torch.optim.Optimizer,
                    device: torch.device, epoch: int, loss_scaler, max_norm: float = 0,
                    mixup_fn: Optional[Mixup] = None, log_writer=None,
                    args=None):
    model.train(True)
    metric_logger = misc.MetricLogger(delimiter="  ")
    metric_logger.add_meter('lr', misc.SmoothedValue(window_size=1, fmt='{value:.6f}'))
    header = 'Epoch: [{}]'.format(epoch)
    print_freq = 20

    accum_iter = args.accum_iter

    optimizer.zero_grad()

    if log_writer is not None:
        print('log_dir: {}'.format(log_writer.log_dir))

    for data_iter_step, (samples, targets) in enumerate(metric_logger.log_every(data_loader, print_freq, header)):

        # we use a per iteration (instead of per epoch) lr scheduler
        if data_iter_step % accum_iter == 0:
            adjust_learning_rate(optimizer, data_iter_step / len(data_loader) + epoch, args)

        samples = samples.to(device, non_blocking=True)
        targets = targets.to(device, non_blocking=True)

        if mixup_fn is not None:
            samples, targets = mixup_fn(samples, targets)

        with torch.cuda.amp.autocast():
            outputs = model(samples)
            loss = criterion(outputs, targets)

        loss_value = loss.item()

        if not math.isfinite(loss_value):
            print("Loss is {}, stopping training".format(loss_value))
            sys.exit(1)

        loss /= accum_iter
        loss_scaler(loss, optimizer, clip_grad=max_norm,
                    parameters=model.parameters(), create_graph=False,
                    update_grad=(data_iter_step + 1) % accum_iter == 0)
        if (data_iter_step + 1) % accum_iter == 0:
            optimizer.zero_grad()

        torch.cuda.synchronize()

        metric_logger.update(loss=loss_value)
        min_lr = 10.
        max_lr = 0.
        for group in optimizer.param_groups:
            min_lr = min(min_lr, group["lr"])
            max_lr = max(max_lr, group["lr"])

        metric_logger.update(lr=max_lr)

        loss_value_reduce = misc.all_reduce_mean(loss_value)
        if log_writer is not None and (data_iter_step + 1) % accum_iter == 0:
            """ We use epoch_1000x as the x-axis in tensorboard.
            This calibrates different curves when batch size changes.
            """
            epoch_1000x = int((data_iter_step / len(data_loader) + epoch) * 1000)
            log_writer.add_scalar('loss', loss_value_reduce, epoch_1000x)
            log_writer.add_scalar('lr', max_lr, epoch_1000x)

    # gather the stats from all processes
    metric_logger.synchronize_between_processes()
    print("Averaged stats:", metric_logger)
    return {k: meter.global_avg for k, meter in metric_logger.meters.items()}


@torch.no_grad()
def evaluate(data_loader, model, device):
    criterion = torch.nn.CrossEntropyLoss()

    metric_logger = misc.MetricLogger(delimiter="  ")
    header = 'Test:'

    # switch to evaluation mode
    model.eval()

    for batch in metric_logger.log_every(data_loader, 10, header):
        images = batch[0]
        target = batch[-1]
        images = images.to(device, non_blocking=True)
        target = target.to(device, non_blocking=True)

        # compute output
        with torch.cuda.amp.autocast():
            output = model(images)
            loss = criterion(output, target)

        acc1, acc5 = accuracy(output, target, topk=(1, 5))

        batch_size = images.shape[0]
        metric_logger.update(loss=loss.item())
        metric_logger.meters['acc1'].update(acc1.item(), n=batch_size)
        metric_logger.meters['acc5'].update(acc5.item(), n=batch_size)
    # gather the stats from all processes
    metric_logger.synchronize_between_processes()
    print('* Acc@1 {top1.global_avg:.3f} Acc@5 {top5.global_avg:.3f} loss {losses.global_avg:.3f}'
          .format(top1=metric_logger.acc1, top5=metric_logger.acc5, losses=metric_logger.loss))

    return {k: meter.global_avg for k, meter in metric_logger.meters.items()}


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

    dataset_train = build_dataset(is_train=True, args=args)
    dataset_val = build_dataset(is_train=False, args=args)

    if (args.subset is not None) and (args.subset >= 1):
        dataset_train = data_utils.subset_of_ImageNet_train_split(dataset_train, args.subset)

    if True:  # args.distributed:
        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))
        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=True)  # shuffle=True to reduce monitor bias
        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)

    os.makedirs(args.output_dir, exist_ok=True)
    if global_rank == 0 and not args.eval:
        args.log_dir = os.path.join(args.output_dir, "tensorboard")
        os.makedirs(args.log_dir, exist_ok=True)
        log_writer = SummaryWriter(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,
    )

    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=False
    )

    print(f"Data loaded for training: there are {len(dataset_train)} images.")
    print(f"Data loaded for validation: there are {len(dataset_val)} images.")

    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 = models_vit.__dict__[args.model](
        num_classes=args.nb_classes,
        drop_path_rate=args.drop_path,
        global_pool=args.global_pool,
        fc_norm=args.use_fc_norm)

    if args.finetune and not args.eval:
        load_pretrained_encoder(model, args.finetune, 
            use_teacher=args.use_teacher, avg_pooling=args.global_pool,
            use_fc_norm=args.use_fc_norm, no_linear_head=False)
        # manually initialize fc layer
        trunc_normal_(model.head.weight, std=2e-5)

    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

    # build optimizer with layer-wise lr decay (lrd)
    param_groups = param_groups_lrd(model_without_ddp, args.weight_decay,
        no_weight_decay_list=model_without_ddp.no_weight_decay(),
        layer_decay=args.layer_decay
    )
    optimizer = torch.optim.AdamW(param_groups, lr=args.lr)
    loss_scaler = NativeScaler()

    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.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,
            args.clip_grad, mixup_fn,
            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, filename="checkpoint-last")
            if (epoch % 10 == 0 or (epoch + 1) == args.epochs):
                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}%')

        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)