Data

Data

timm.data.create_dataset

<source>

( namerootsplit = 'validation'search_split = Trueclass_map = Noneload_bytes = Falseis_training = Falsedownload = Falsebatch_size = Noneseed = 42repeats = 0**kwargs )

Dataset factory method

In parenthesis after each arg are the type of dataset supported for each arg, one of:

• folder - default, timm folder (or tar) based ImageDataset

• torch - torchvision based datasets

• HFDS - Hugging Face Datasets

• TFDS - Tensorflow-datasets wrapper in IterabeDataset interface via IterableImageDataset

• WDS - Webdataset

• all - any of the above

timm.data.create_loader

<source>

( datasetinput_sizebatch_sizeis_training = Falseuse_prefetcher = Trueno_aug = Falsere_prob = 0.0re_mode = 'const're_count = 1re_split = Falsescale = Noneratio = Nonehflip = 0.5vflip = 0.0color_jitter = 0.4auto_augment = Nonenum_aug_repeats = 0num_aug_splits = 0interpolation = 'bilinear'mean = (0.485, 0.456, 0.406)std = (0.229, 0.224, 0.225)num_workers = 1distributed = Falsecrop_pct = Nonecrop_mode = Nonecollate_fn = Nonepin_memory = Falsefp16 = Falseimg_dtype = torch.float32device = device(type='cuda')tf_preprocessing = Falseuse_multi_epochs_loader = Falsepersistent_workers = Trueworker_seeding = 'all' )

timm.data.create_transform

<source>

( input_sizeis_training = Falseuse_prefetcher = Falseno_aug = Falsescale = Noneratio = Nonehflip = 0.5vflip = 0.0color_jitter = 0.4auto_augment = Noneinterpolation = 'bilinear'mean = (0.485, 0.456, 0.406)std = (0.229, 0.224, 0.225)re_prob = 0.0re_mode = 'const're_count = 1re_num_splits = 0crop_pct = Nonecrop_mode = Nonetf_preprocessing = Falseseparate = False )

timm.data.resolve_data_config

<source>

( args = Nonepretrained_cfg = Nonemodel = Noneuse_test_size = Falseverbose = False )