trainval_ce.py

from __future__ import print_function

import os
import sys
import argparse
import time
import math

import tensorboard_logger as tb_logger
import torch
import torch.backends.cudnn as cudnn
from torchvision import transforms, datasets
from dataloader.cubs2011 import CUBS
from dataloader.imagenet import ImageNet

from util import AverageMeter, scale_255, list_of_ints
from util import adjust_learning_rate, warmup_learning_rate, accuracy
from util import set_optimizer, save_model
from networks.resnet_big import SupCEResNet

# Imbalanced dataset creation
import numpy as np
import random

# MedMNIST dataset
import medmnist
from medmnist import INFO # INFO allows us to fetch the details of dataset based on search flag

try:
    import apex
    from apex import amp, optimizers
except ImportError:
    pass


def parse_option():
    parser = argparse.ArgumentParser('argument for training')

    parser.add_argument('--print_freq', type=int, default=10,
                        help='print frequency')
    parser.add_argument('--save_freq', type=int, default=50,
                        help='save frequency')
    parser.add_argument('--batch_size', type=int, default=256,
                        help='batch_size')
    parser.add_argument('--num_workers', type=int, default=16,
                        help='num of workers to use')
    parser.add_argument('--epochs', type=int, default=500,
                        help='number of training epochs')

    # optimization
    parser.add_argument('--learning_rate', type=float, default=0.2,
                        help='learning rate')
    parser.add_argument('--lr_decay_epochs', type=str, default='350,400,450',
                        help='where to decay lr, can be a list')
    parser.add_argument('--lr_decay_rate', type=float, default=0.1,
                        help='decay rate for learning rate')
    parser.add_argument('--weight_decay', type=float, default=1e-4,
                        help='weight decay')
    parser.add_argument('--momentum', type=float, default=0.9,
                        help='momentum')

    # model dataset
    parser.add_argument('--model', type=str, default='resnet50')
    parser.add_argument('--dataset', type=str, default='cifar10',
                        choices=['cifar10', 'organamnist', 'dermamnist', 'bloodmnist', 'cifar100', 'cubs', 'imagenet', 'imagenet32'], help='dataset')

    # class imbalance parameters
    parser.add_argument('--use_imbalanced', action='store_true',
                        help='using Class Imbalanced dataset')
    parser.add_argument('--imbalance_ratio', type=int, default=10,
                        help='Imbalance ratio for imbalanced data sampling')
    parser.add_argument('--imbalance_classes', type=list_of_ints, default=None,
                        help='List of class IDs in the range of [0, max_no_of_classes]')

    # other setting
    parser.add_argument('--cosine', action='store_true',
                        help='using cosine annealing')
    parser.add_argument('--constant', action='store_true',
                        help='using fixed LR')
    parser.add_argument('--syncBN', action='store_true',
                        help='using synchronized batch normalization')
    parser.add_argument('--warm', action='store_true',
                        help='warm-up for large batch training')
    parser.add_argument('--trial', type=str, default='0',
                        help='id for recording multiple runs')

    opt = parser.parse_args()

    # set the path according to the environment
    opt.data_folder = './datasets/'
    opt.model_path = './save/SupCon/{}_models'.format(opt.dataset)
    opt.tb_path = './save/SupCon/{}_tensorboard'.format(opt.dataset)

    iterations = opt.lr_decay_epochs.split(',')
    opt.lr_decay_epochs = list([])
    for it in iterations:
        opt.lr_decay_epochs.append(int(it))

    opt.model_name = 'SupCE_{}_{}_lr_{}_decay_{}_bsz_{}_trial_{}'.\
        format(opt.dataset, opt.model, opt.learning_rate, opt.weight_decay,
               opt.batch_size, opt.trial)

    if opt.cosine:
        opt.model_name = '{}_cosine'.format(opt.model_name)

    if opt.use_imbalanced:
        opt.model_name = '{}_imbalanced_{}_ratio'.format(opt.model_name, opt.imbalance_ratio)

    # warm-up for large-batch training,
    if opt.batch_size > 256:
        opt.warm = True
    if opt.warm:
        opt.model_name = '{}_warm'.format(opt.model_name)
        opt.warmup_from = 0.01
        opt.warm_epochs = 10
        if opt.cosine:
            eta_min = opt.learning_rate * (opt.lr_decay_rate ** 3)
            opt.warmup_to = eta_min + (opt.learning_rate - eta_min) * (
                    1 + math.cos(math.pi * opt.warm_epochs / opt.epochs)) / 2
        else:
            opt.warmup_to = opt.learning_rate

    opt.tb_folder = os.path.join(opt.tb_path, opt.model_name)
    if not os.path.isdir(opt.tb_folder):
        os.makedirs(opt.tb_folder)

    opt.save_folder = os.path.join(opt.model_path, opt.model_name)
    if not os.path.isdir(opt.save_folder):
        os.makedirs(opt.save_folder)

    if opt.dataset == 'cifar10':
        opt.n_cls = 10
    elif opt.dataset == 'organamnist':
        opt.n_cls = 11
    elif opt.dataset == 'dermamnist':
        opt.n_cls = 7
    elif opt.dataset == 'bloodmnist':
        opt.n_cls = 8
    elif opt.dataset == 'cifar100':
        opt.n_cls = 100
    elif opt.dataset == 'cubs':
        opt.n_cls = 200
    elif opt.dataset == 'imagenet':
        opt.n_cls = 1000
    elif opt.dataset == 'imagenet32':
        opt.n_cls = 1000
    else:
        raise ValueError('dataset not supported: {}'.format(opt.dataset))

    return opt

def create_class_imbal_indices(dataset, imbalance_ratio, class_set_idx=None):
    # Get all training targets and count the number of class instances
    targets = np.array(dataset.targets)
    classes, class_counts = np.unique(targets, return_counts=True)
    nb_classes = len(classes)
    
    # compute the imbalanced class counts
    # chose between step function or longtail
    if class_set_idx is None:
        # longtail dist
        mu = np.power(1 / imbalance_ratio, 1 / (nb_classes - 1))
        imbal_class_counts = [int(max(class_counts) * np.power(mu, i)) for i in range(nb_classes)]
    else:
        # Step function
        imbal_class_counts = np.ones(nb_classes, dtype=int) * int(max(class_counts))
        for i in class_set_idx:
            imbal_class_counts[i] = int(imbal_class_counts[i] * (1 / imbalance_ratio))
    imbal_class_counts = list(imbal_class_counts)
    print(imbal_class_counts)

    # get balanced class indices
    class_indices = [np.where(targets == i)[0] for i in range(nb_classes)]

    # sample for imbalanced distribution
    imbal_class_indices = [random.sample(list(class_idx), class_count) \
                           for class_idx, class_count in \
                            zip(class_indices, imbal_class_counts)]
    
    imbal_class_indices = np.hstack(imbal_class_indices)

    return imbal_class_indices

def set_loader(opt):
    # construct data loader
    if opt.dataset == 'cifar10':
        mean = (0.4914, 0.4822, 0.4465)
        std = (0.2023, 0.1994, 0.2010)
        
        normalize = transforms.Normalize(mean=mean, std=std)
        
        train_transform = transforms.Compose([
            transforms.RandomResizedCrop(size=32, scale=(0.2, 1.)),
            transforms.RandomHorizontalFlip(),
            transforms.ToTensor(),
            normalize,
        ])
    elif opt.dataset[-5:] == 'mnist': # this is a single channel dataset
        opt.data_folder = os.path.join(opt.data_folder, "medmnist")
        mean = (0.5)
        std = (0.5)
        
        normalize = transforms.Normalize(mean=mean, std=std)
        train_transform = transforms.Compose([
            transforms.RandomResizedCrop(size=28, scale=(0.2, 1.)),
            transforms.RandomHorizontalFlip(),
            transforms.ToTensor(),
            normalize,
        ])
    elif opt.dataset == 'cifar100':
        mean = (0.5071, 0.4867, 0.4408)
        std = (0.2675, 0.2565, 0.2761)
        
        normalize = transforms.Normalize(mean=mean, std=std)
        train_transform = transforms.Compose([
            transforms.RandomResizedCrop(size=32, scale=(0.2, 1.)),
            transforms.RandomHorizontalFlip(),
            transforms.ToTensor(),
            normalize,
        ])
    elif opt.dataset == 'cubs':
        mean = (123., 117., 104.)
        std = (1., 1., 1.)
        
        normalize = transforms.Normalize(mean=mean, std=std)
        train_transform = transforms.Compose([
            transforms.Resize(256),
            transforms.RandomResizedCrop(size=224),
            transforms.RandomHorizontalFlip(),
            transforms.ToTensor(),
            scale_255(),
            normalize
        ])
    elif opt.dataset == 'imagenet':
        mean = (0.485, 0.456, 0.406)
        std = (0.229, 0.224, 0.225)
        
        normalize = transforms.Normalize(mean=mean, std=std)
        train_transform = transforms.Compose([
            transforms.RandomResizedCrop(size=224, scale=(0.2, 1.)),
            transforms.RandomHorizontalFlip(),
            transforms.ToTensor(),
            normalize,
        ])
    else:
        raise ValueError('dataset not supported: {}'.format(opt.dataset))

    val_transform = transforms.Compose([
        transforms.ToTensor(),
        normalize,
    ])
    if opt.dataset == 'cubs':
        val_transform = transforms.Compose([
            transforms.Resize(size=(224,224)),
            transforms.ToTensor(),
            transforms.Normalize(mean=(0.485, 0.456, 0.406), 
                                std=(0.229, 0.224, 0.225)),
        ])

    if opt.dataset == 'cifar10':
        train_dataset = datasets.CIFAR10(root=opt.data_folder,
                                         transform=train_transform,
                                         download=True)
        val_dataset = datasets.CIFAR10(root=opt.data_folder,
                                       train=False,
                                       transform=val_transform)
    elif opt.dataset[-5:] == 'mnist':
        info = INFO[opt.dataset]
        print(info)
        DataClass = getattr(medmnist, info['python_class'])

        train_dataset = DataClass(root=opt.data_folder,
                                split="train",
                                transform=train_transform,
                                as_rgb=True,
                                download=False)
        val_dataset = DataClass(root=opt.data_folder,
                                split='test', 
                                transform=val_transform, 
                                download=False, 
                                as_rgb=True)
    elif opt.dataset == 'cifar100':
        train_dataset = datasets.CIFAR100(root=opt.data_folder,
                                          transform=train_transform,
                                          download=True)
        val_dataset = datasets.CIFAR100(root=opt.data_folder,
                                        train=False,
                                        transform=val_transform)
    elif opt.dataset == 'cubs':
        train_dataset = CUBS(root=opt.data_folder, train = True,
                                            transform=train_transform)
        val_dataset = CUBS(root=opt.data_folder, train = False,
                                            transform=val_transform)
    elif opt.dataset == 'imagenet':
        train_dataset = ImageNet(root=opt.data_folder, split='train',
                                 transform=train_transform)
        val_dataset = ImageNet(root=opt.data_folder, split='val',
                               transform=val_transform)
    else:
        raise ValueError(opt.dataset)

    if opt.use_imbalanced:
        imbal_class_idx = create_class_imbal_indices(train_dataset, opt.imbalance_ratio, opt.imbalance_classes)
        train_dataset.data = train_dataset.data[imbal_class_idx]
        train_dataset.targets = np.array(train_dataset.targets)[imbal_class_idx]

        assert len(train_dataset.targets) == len(train_dataset.data)

    train_sampler = None
    train_loader = torch.utils.data.DataLoader(
        train_dataset, batch_size=opt.batch_size, shuffle=(train_sampler is None),
        num_workers=opt.num_workers, pin_memory=True, sampler=train_sampler)
    val_loader = torch.utils.data.DataLoader(
        val_dataset, batch_size=256, shuffle=False,
        num_workers=8, pin_memory=True)

    return train_loader, val_loader


def set_model(opt):
    model = SupCEResNet(name=opt.model, num_classes=opt.n_cls)
    criterion = torch.nn.CrossEntropyLoss()

    # enable synchronized Batch Normalization
    if opt.syncBN:
        model = apex.parallel.convert_syncbn_model(model)

    if torch.cuda.is_available():
        if torch.cuda.device_count() > 1:
            model = torch.nn.DataParallel(model)
        model = model.cuda()
        criterion = criterion.cuda()
        cudnn.benchmark = True

    return model, criterion


def train(train_loader, model, criterion, optimizer, epoch, opt):
    """one epoch training"""
    model.train()

    batch_time = AverageMeter()
    data_time = AverageMeter()
    losses = AverageMeter()
    top1 = AverageMeter()

    end = time.time()
    for idx, (images, labels) in enumerate(train_loader):
        data_time.update(time.time() - end)

        images = images.cuda(non_blocking=True)
        labels = labels.cuda(non_blocking=True)
        bsz = labels.shape[0]

        # warm-up learning rate
        warmup_learning_rate(opt, epoch, idx, len(train_loader), optimizer)

        # compute loss
        output = model(images)
        labels = labels.view(-1)
        loss = criterion(output, labels)

        # update metric
        losses.update(loss.item(), bsz)
        acc1, acc5 = accuracy(output, labels, topk=(1, 5))
        top1.update(acc1[0], bsz)

        # SGD
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        # measure elapsed time
        batch_time.update(time.time() - end)
        end = time.time()

        # print info
        if (idx + 1) % opt.print_freq == 0:
            print('Train: [{0}][{1}/{2}]\t'
                  'BT {batch_time.val:.3f} ({batch_time.avg:.3f})\t'
                  'DT {data_time.val:.3f} ({data_time.avg:.3f})\t'
                  'loss {loss.val:.3f} ({loss.avg:.3f})\t'
                  'Acc@1 {top1.val:.3f} ({top1.avg:.3f})'.format(
                   epoch, idx + 1, len(train_loader), batch_time=batch_time,
                   data_time=data_time, loss=losses, top1=top1))
            sys.stdout.flush()

    return losses.avg, top1.avg


def validate(val_loader, model, criterion, opt):
    """validation"""
    model.eval()

    batch_time = AverageMeter()
    losses = AverageMeter()
    top1 = AverageMeter()

    with torch.no_grad():
        end = time.time()
        for idx, (images, labels) in enumerate(val_loader):
            images = images.float().cuda()
            labels = labels.cuda()
            bsz = labels.shape[0]

            # forward
            output = model(images)
            labels = labels.view(-1)
            loss = criterion(output, labels)

            # update metric
            losses.update(loss.item(), bsz)
            acc1, acc5 = accuracy(output, labels, topk=(1, 5))
            top1.update(acc1[0], bsz)

            # measure elapsed time
            batch_time.update(time.time() - end)
            end = time.time()

            if idx % opt.print_freq == 0:
                print('Test: [{0}/{1}]\t'
                      'Time {batch_time.val:.3f} ({batch_time.avg:.3f})\t'
                      'Loss {loss.val:.4f} ({loss.avg:.4f})\t'
                      'Acc@1 {top1.val:.3f} ({top1.avg:.3f})'.format(
                       idx, len(val_loader), batch_time=batch_time,
                       loss=losses, top1=top1))

    print(' * Acc@1 {top1.avg:.3f}'.format(top1=top1))
    return losses.avg, top1.avg


def main():
    best_acc = 0
    opt = parse_option()

    # build data loader
    train_loader, val_loader = set_loader(opt)

    # build model and criterion
    model, criterion = set_model(opt)

    # build optimizer
    optimizer = set_optimizer(opt, model)

    # tensorboard
    logger = tb_logger.Logger(logdir=opt.tb_folder, flush_secs=2)

    # training routine
    for epoch in range(1, opt.epochs + 1):
        adjust_learning_rate(opt, optimizer, epoch)

        # train for one epoch
        time1 = time.time()
        loss, train_acc = train(train_loader, model, criterion, optimizer, epoch, opt)
        time2 = time.time()
        print('epoch {}, total time {:.2f}'.format(epoch, time2 - time1))

        # tensorboard logger
        logger.log_value('train_loss', loss, epoch)
        logger.log_value('train_acc', train_acc, epoch)
        logger.log_value('learning_rate', optimizer.param_groups[0]['lr'], epoch)

        # evaluation
        loss, val_acc = validate(val_loader, model, criterion, opt)
        logger.log_value('val_loss', loss, epoch)
        logger.log_value('val_acc', val_acc, epoch)

        if val_acc > best_acc:
            best_acc = val_acc

        if epoch % opt.save_freq == 0:
            save_file = os.path.join(
                opt.save_folder, 'ckpt_epoch_{epoch}.pth'.format(epoch=epoch))
            save_model(model, optimizer, opt, epoch, save_file)

    # save the last model
    save_file = os.path.join(
        opt.save_folder, 'last.pth')
    save_model(model, optimizer, opt, opt.epochs, save_file)

    print('best accuracy: {:.2f}'.format(best_acc))


if __name__ == '__main__':
    main()