data.datasets package

Subpackages

• data.datasets.audio_classification package
  • Submodules
  • data.datasets.audio_classification.speech_commands_v2 module
    • SpeechCommandsv2Dataset
      • SpeechCommandsv2Dataset.__init__()
      • SpeechCommandsv2Dataset.add_arguments()
      • SpeechCommandsv2Dataset.get_sample()
      • SpeechCommandsv2Dataset.get_transformed_sample()
  • Module contents
• data.datasets.classification package
  • Submodules
  • data.datasets.classification.base_image_classification_dataset module
    • BaseImageClassificationDataset
      • BaseImageClassificationDataset.__init__()
      • BaseImageClassificationDataset.add_arguments()
      • BaseImageClassificationDataset.share_dataset_arguments()
      • BaseImageClassificationDataset.extra_repr()
  • data.datasets.classification.base_imagenet_shift_dataset module
    • BaseImageNetShiftDataset
      • BaseImageNetShiftDataset.__init__()
      • BaseImageNetShiftDataset.post_init_checks()
      • BaseImageNetShiftDataset.class_id_to_imagenet_class_id()
  • data.datasets.classification.imagenet module
    • ImageNetDataset
      • ImageNetDataset.__init__()
  • data.datasets.classification.imagenet_a module
    • ImageNetADataset
      • ImageNetADataset.__init__()
      • ImageNetADataset.class_id_to_imagenet_class_id()
  • data.datasets.classification.imagenet_r module
    • ImageNetRDataset
      • ImageNetRDataset.__init__()
      • ImageNetRDataset.class_id_to_imagenet_class_id()
  • data.datasets.classification.imagenet_sketch module
    • ImageNetSketchDataset
      • ImageNetSketchDataset.__init__()
      • ImageNetSketchDataset.class_id_to_imagenet_class_id()
  • data.datasets.classification.imagenet_synsets module
  • data.datasets.classification.imagenet_v2 module
    • Imagenetv2Dataset
      • Imagenetv2Dataset.__init__()
      • Imagenetv2Dataset.add_arguments()
  • data.datasets.classification.places365 module
    • Places365Dataset
      • Places365Dataset.__init__()
  • Module contents
• data.datasets.detection package
  • Submodules
  • data.datasets.detection.base_detection module
    • BaseDetectionDataset
      • BaseDetectionDataset.__init__()
      • BaseDetectionDataset.add_arguments()
  • data.datasets.detection.coco_base module
    • COCODetection
      • COCODetection.__init__()
      • COCODetection.share_dataset_arguments()
      • COCODetection.add_arguments()
      • COCODetection.get_boxes_and_labels()
      • COCODetection.get_image()
      • COCODetection.extra_repr()
      • COCODetection.class_names()
  • data.datasets.detection.coco_mask_rcnn module
    • COCODetectionMaskRCNN
      • COCODetectionMaskRCNN.__init__()
      • COCODetectionMaskRCNN.add_arguments()
    • coco_mask_rcnn_collate_fn()
  • data.datasets.detection.coco_ssd module
    • COCODetectionSSD
      • COCODetectionSSD.__init__()
      • COCODetectionSSD.generate_anchors()
      • COCODetectionSSD.extra_repr()
    • coco_ssd_collate_fn()
  • Module contents
• data.datasets.multi_modal_img_text package
  • Subpackages
    • data.datasets.multi_modal_img_text.zero_shot package
      • Submodules
      • data.datasets.multi_modal_img_text.zero_shot.base_zero_shot module
      • data.datasets.multi_modal_img_text.zero_shot.imagenet module
      • data.datasets.multi_modal_img_text.zero_shot.imagenet_class_names module
      • data.datasets.multi_modal_img_text.zero_shot.templates module
      • Module contents
  • Submodules
  • data.datasets.multi_modal_img_text.base_multi_modal_img_text module
    • BaseMultiModalImgText
      • BaseMultiModalImgText.__init__()
      • BaseMultiModalImgText.get_zero_shot_dataset()
      • BaseMultiModalImgText.get_dataset()
      • BaseMultiModalImgText.share_dataset_arguments()
      • BaseMultiModalImgText.add_arguments()
      • BaseMultiModalImgText.get_zero_shot_pair()
      • BaseMultiModalImgText.get_dataset_pair()
      • BaseMultiModalImgText.extra_repr()
    • multi_modal_img_text_collate_fn()
  • data.datasets.multi_modal_img_text.flickr module
    • FlickrDataset
      • FlickrDataset.get_dataset()
  • data.datasets.multi_modal_img_text.img_text_tar_dataset module
    • extract_content()
    • decode_image()
    • decode_text()
    • async_download_file_from_s3()
    • ImgTextTarDataset
      • ImgTextTarDataset.__init__()
      • ImgTextTarDataset.get_dataset()
      • ImgTextTarDataset.add_arguments()
      • ImgTextTarDataset.get_dataset_pair()
  • Module contents
    • arguments_multi_modal_img_text()
• data.datasets.segmentation package
  • Submodules
  • data.datasets.segmentation.ade20k module
    • ADE20KDataset
      • ADE20KDataset.__init__()
      • ADE20KDataset.adjust_mask_value()
      • ADE20KDataset.color_palette()
      • ADE20KDataset.class_names()
  • data.datasets.segmentation.base_segmentation module
    • BaseImageSegmentationDataset
      • BaseImageSegmentationDataset.__init__()
      • BaseImageSegmentationDataset.add_arguments()
      • BaseImageSegmentationDataset.check_dataset()
      • BaseImageSegmentationDataset.adjust_mask_value()
      • BaseImageSegmentationDataset.color_palette()
      • BaseImageSegmentationDataset.class_names()
      • BaseImageSegmentationDataset.read_mask_pil()
      • BaseImageSegmentationDataset.convert_mask_to_tensor()
  • data.datasets.segmentation.coco_segmentation module
    • COCOSegmentationDataset
      • COCOSegmentationDataset.__init__()
      • COCOSegmentationDataset.class_names()
      • COCOSegmentationDataset.coco_to_pascal_mapping()
  • data.datasets.segmentation.pascal_voc module
    • PascalVOCDataset
      • PascalVOCDataset.__init__()
      • PascalVOCDataset.add_arguments()
      • PascalVOCDataset.color_palette()
      • PascalVOCDataset.class_names()
  • Module contents
• data.datasets.utils package
  • Submodules
  • data.datasets.utils.common module
    • file_has_valid_image_extension()
    • file_has_allowed_extension()
    • get_image_paths()
    • select_random_subset()
    • select_samples_by_category()
  • data.datasets.utils.text module
    • caption_preprocessing()
  • data.datasets.utils.video module
    • fetch_labels_from_timestamps()
  • Module contents

Submodules

data.datasets.dataset_base module

class data.datasets.dataset_base.BaseDataset(opts: Namespace, is_training: bool = True, is_evaluation: bool = False, *args, **kwargs)

Bases: Dataset, ABC

Base class for creating datasets. Sub-classes must implement __getitem__, _training_transforms, and _validation_transforms functions.

Parameters:

• opts – Command-line arguments

• is_training – Training mode or not. Defaults to True.

• is_evaluation – Evaluation mode or not. Defaults to False.

…note::

is_training is used to indicate whether the dataset is used for training or validation. On the other hand, is_evaluation mode is used to indicate the dataset is used for testing.

Theoretically, is_training=False and is_evaluation=True should be the same. However, for some datasets (especially segmentation), validation dataset transforms are different from test transforms because each image has different resolution, making it difficult to construct batches. Therefore, we treat these two modes different.

For datasets, where validation and testing transforms are the same, we set evaluation transforms the same as the validation transforms.

__init__(opts: Namespace, is_training: bool = True, is_evaluation: bool = False, *args, **kwargs) → None

classmethod add_arguments(parser: ArgumentParser) → ArgumentParser

Add dataset-specific arguments

static load_from_server(opts: Namespace, is_training: bool, is_evaluation: bool) → Namespace | None

Helper function to load dataset from server.

get_augmentation_transforms(*args, **kwargs) → BaseTransformation

Helper function to get data transforms depending on the mode (training, evaluation, or validation)

share_dataset_arguments() → Dict[str, Any]

Function that can be used by sub-classes to share dataset-specific options. It returns a mapping. An example is {“model.classification.n_classes”, 1000}

By default, we return an empty dictionary

extra_repr() → str

Extra information to be represented in __repr__. Each line in the output string should be prefixed with \t.

get_item_metadata(item_idx: int) → Dict

Returns the metadata for given @item_idx. This method could be used by samplers for sampling dynamic batches based on the metadata of the items.

Parameters:

item_idx (int) – The index of sample to provide metadata for. The indexing should be aligned with how self.__getitem__(item_idx) sequences the dataset items.

Returns: A dict containing the metadata. Each sampler may require a specific

schema to be returned by this function.

class data.datasets.dataset_base.BaseImageDataset(opts: Namespace, is_training: bool = True, is_evaluation: bool = False, *args, **kwargs)

Bases: BaseDataset, ABC

Base Dataset class for Image datasets.

static read_image_pil(path: str) → Image | None

Reads a PIL image.

Parameters:

path – Path of image file.

Returns:

If there are no exceptions (e.g., because of corrupted images), PIL Image is returned. Otherwise, None.

extra_repr() → str

Extra information to be represented in __repr__. Each line in the output string should be prefixed with \t.

class data.datasets.dataset_base.BaseVideoDataset(opts: Namespace, *args, **kwargs)

Bases: BaseDataset, ABC

Base Dataset class for video datasets.

Parameters:

opts – Command-line arguments

__init__(opts: Namespace, *args, **kwargs) → None

property clips_per_video: int

property n_frames_per_clip: int

classmethod add_arguments(parser: ArgumentParser) → ArgumentParser

Add dataset-specific arguments

get_item_metadata(item_idx: int) → VideoMetadataDict

Returns the metadata for given @item_idx. This method is used by VideoClipSampler for sampling dynamic clips based on the duration of the items.

Subclasses should override this method if they use VideoClipSampler.

Parameters:

item_idx (int) – The index of video file to provide metadata. The indexing should be aligned with how self.__getitem__(item_idx) sequences the dataset items.

Returns: A dict containing the metadata. Please see @VideoMetadataDict documentation.

class data.datasets.dataset_base.VideoMetadataDict

Bases: TypedDict

This class is an alias of Dict, in addition to optional standard key names and value types. The fields will be required or optional depending on the Sampler.

video_fps: float

total_video_frames: int

video_duration: float

audio_fps: float

Module contents

data.datasets.build_dataset_from_registry(opts: Namespace, is_training: bool = True, is_evaluation: bool = False, *args, **kwargs) → BaseDataset

Helper function to build a dataset from dataset registry

Parameters:

• opts – Command-line arguments

• is_training – Training mode or not. Defaults to True.

• is_evaluation – Evaluation mode or not. Defaults to False.

Returns:

An instance of BaseDataset

…note:

is_training is used to indicate whether the dataset is used for training or validation On the other hand, is_evaluation mode is used to indicate the dataset is used for testing.

Theoretically, is_training=False and is_evaluation=True should be the same. However, for some datasets (especially segmentation), validation dataset transforms are different from test transforms because each image has different resolution, making it difficult to construct batches. Therefore, we treat these two modes different. For datasets, where validation and testing transforms are the same, we set evaluation transforms the same as the validation transforms (e.g., in ImageNet object classification).

data.datasets.get_test_dataset(opts: Namespace, *args, **kwargs) → BaseDataset

Helper function to build a dataset for testing.

Parameters:

opts – Command-line arguments

Returns:

An instance of BaseDataset

data.datasets.get_train_val_datasets(opts: Namespace, *args, **kwargs) → Tuple[BaseDataset, BaseDataset | None]

Helper function to build a dataset for training and validation.

Parameters:

opts – Command-line arguments

Returns:

Training and (optionally) validation datasets.

data.datasets.arguments_dataset(parser: ArgumentParser) → ArgumentParser

Add dataset-specific arguments from BaseDataset, BaseImageDataset, BaseImageClassificationDataset, BaseImageNetShiftDataset, BaseVideoDataset, zero-shot datasets, and DATASET_REGISTRY.