datasets.py

# Copyright (c) 2015-present, Facebook, Inc.
# All rights reserved.
import os
import json

from torchvision import datasets, transforms
from torchvision.datasets.folder import ImageFolder, default_loader
from torch.utils.data import Subset, ConcatDataset

from timm.data.constants import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
from timm.data import create_transform
import numpy as np
from torchvision.datasets import ImageFolder
from PIL import Image
import torch

SUBSET_SIZE=100
IMAGENET_DEFAULT_MEAN = (0.485, 0.456, 0.406)
IMAGENET_DEFAULT_STD = (0.229, 0.224, 0.225)
CIFAR10_MEAN = (0.4914, 0.4822, 0.4465)
CIFAR10_STD = (0.2471, 0.2435, 0.2616)
CIFAR100_MEAN = (0.5071, 0.4867, 0.4408)
CIFAR100_STD = (0.2675, 0.2565, 0.2761)
class INatDataset(ImageFolder):
    def __init__(self, root, train=True, year=2018, transform=None, target_transform=None,
                 category='name', loader=default_loader):
        self.transform = transform
        self.loader = loader
        self.target_transform = target_transform
        self.year = year
        # assert category in ['kingdom','phylum','class','order','supercategory','family','genus','name']
        path_json = os.path.join(root, f'{"train" if train else "val"}{year}.json')
        with open(path_json) as json_file:
            data = json.load(json_file)

        with open(os.path.join(root, 'categories.json')) as json_file:
            data_catg = json.load(json_file)

        path_json_for_targeter = os.path.join(root, f"train{year}.json")

        with open(path_json_for_targeter) as json_file:
            data_for_targeter = json.load(json_file)

        targeter = {}
        indexer = 0
        for elem in data_for_targeter['annotations']:
            king = []
            king.append(data_catg[int(elem['category_id'])][category])
            if king[0] not in targeter.keys():
                targeter[king[0]] = indexer
                indexer += 1
        self.nb_classes = len(targeter)

        self.samples = []
        for elem in data['images']:
            cut = elem['file_name'].split('/')
            target_current = int(cut[2])
            path_current = os.path.join(root, cut[0], cut[2], cut[3])

            categors = data_catg[target_current]
            target_current_true = targeter[categors[category]]
            self.samples.append((path_current, target_current_true))

    # __getitem__ and __len__ inherited from ImageFolder



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

    if args.data_set == 'cifar100':
        dataset = datasets.CIFAR100(args.data_path, train=is_train, transform=transform,  download = True)
        nb_classes = 100
    if args.data_set == 'cifar10':
        dataset = datasets.CIFAR10(args.data_path, train=is_train, transform=transform, download = True)
        nb_classes = 10
    elif args.data_set == 'IMNET':
        root = os.path.join(args.data_path, 'train' if is_train else 'val')
        dataset = datasets.ImageFolder(root, transform=transform)
        nb_classes = 1000
    elif args.data_set == 'IMNET_SUBSET':
        root = os.path.join(args.data_path, 'train' if is_train else 'val')
        dataset =  datasets.ImageFolder(root, transform=transform)
        nb_classes = 1000
        classes = None
        np.random.seed(args.data_seed)
        classes = list(np.random.choice([i for i in range(1000)], size=SUBSET_SIZE))
        print((classes))
        dataset = create_subset_from_dataset(dataset, classes)

    elif args.data_set == 'INAT':
        dataset = INatDataset(args.data_path, train=is_train, year=2018,
                              category=args.inat_category, transform=transform)
        nb_classes = dataset.nb_classes
    elif args.data_set == 'INAT19':
        dataset = INatDataset(args.data_path, train=is_train, year=2019,
                              category=args.inat_category, transform=transform)
        nb_classes = dataset.nb_classes

    return dataset, nb_classes


def create_subset_from_dataset(train_dataset, classes):
    mask = np.isin(np.array(train_dataset.targets), classes)
    indices = np.where(mask)
    train_dataset = Subset(train_dataset, indices[0])
    return train_dataset


def build_transform(is_train, args):
    resize_im = args.input_size > 32
    if is_train:
        # this should always dispatch to transforms_imagenet_train
        if args.data_set == 'cifar10':
            transform = create_transform(
                input_size=args.input_size,
                is_training=True,
                color_jitter=args.color_jitter,
                auto_augment=args.aa,
                interpolation=args.train_interpolation,
                re_prob=args.reprob,
                re_mode=args.remode,
                re_count=args.recount,
                mean=CIFAR10_MEAN,
                std=CIFAR10_STD
            )
        elif args.data_set == 'IMNET' or args.data_set == 'IMNET_SUBSET':
            transform = create_transform(
                input_size=args.input_size,
                is_training=True,
                color_jitter=args.color_jitter,
                auto_augment=args.aa,
                interpolation=args.train_interpolation,
                re_prob=args.reprob,
                re_mode=args.remode,
                re_count=args.recount,
                mean=IMAGENET_DEFAULT_MEAN,
                std=IMAGENET_DEFAULT_STD
            )

        elif args.data_set == 'cifar100':
            transform = create_transform(
                input_size=args.input_size,
                is_training=True,
                color_jitter=args.color_jitter,
                auto_augment=args.aa,
                interpolation=args.train_interpolation,
                re_prob=args.reprob,
                re_mode=args.remode,
                re_count=args.recount,
                mean=CIFAR100_MEAN,
                std=CIFAR100_STD
            )
        if not resize_im:
            # replace RandomResizedCropAndInterpolation with
            # RandomCrop
            transform.transforms[0] = transforms.RandomCrop(
                args.input_size, padding=4)
        return transform

    t = []
    if resize_im:
        size = int((256 / 224) * args.input_size)
        t.append(
            transforms.Resize(size, interpolation=3),  # to maintain same ratio w.r.t. 224 images
        )
        t.append(transforms.CenterCrop(args.input_size))

    t.append(transforms.ToTensor())
    if args.data_set == 'cifar10':
        t.append(transforms.Normalize(CIFAR10_MEAN, CIFAR10_STD))
    elif args.data_set == 'IMNET' or args.data_set == 'IMNET_SUBSET':
        t.append(transforms.Normalize(IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD))
    elif args.data_set == 'cifar100':
        t.append(transforms.Normalize(CIFAR100_MEAN, CIFAR100_STD))
    return transforms.Compose(t)