datasets.py

import torch
import os
from torchvision import datasets, transforms

from timm.data.constants import \
    IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD, IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
from timm.data.transforms import RandomResizedCropAndInterpolation
from timm.data import create_transform

from data.build import build_imagenet_dataset
import random
import numpy as np

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

    print("Transform = ")
    if isinstance(transform, tuple):
        for trans in transform:
            print(" - - - - - - - - - - ")
            for t in trans.transforms:
                print(t)
    else:
        for t in transform.transforms:
            print(t)
    print("---------------------------")
    
    dataset = build_imagenet_dataset(args, is_train, transform)
    nb_classes = 1000

    assert nb_classes == args.nb_classes
    print("Number of the class = %d" % args.nb_classes)

    return dataset, nb_classes


def build_transform(is_train, args):
    resize_im = args.input_size > 32
    imagenet_default_mean_and_std = args.imagenet_default_mean_and_std
    mean = IMAGENET_INCEPTION_MEAN if not imagenet_default_mean_and_std else IMAGENET_DEFAULT_MEAN
    std = IMAGENET_INCEPTION_STD if not imagenet_default_mean_and_std else IMAGENET_DEFAULT_STD

    if args.clip_mean_and_std:
        mean = [0.48145466, 0.4578275, 0.40821073]
        std = [0.26862954, 0.26130258, 0.27577711]

    if is_train:
        # this should always dispatch to transforms_imagenet_train
        try:
            scale = (args.crop_scale, 1)
        except:
            scale = None
        print (f"use crop scale {scale}")
        transform = create_transform(
            input_size=args.input_size,
            is_training=True,
            scale=scale,
            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=mean,
            std=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:
        if args.crop_pct is None:
            args.crop_pct = 1.0
        size = int(args.input_size / args.crop_pct)
        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())
    t.append(transforms.Normalize(mean, std))
    return transforms.Compose(t)