def Ptvssv2()

in slowfast/datasets/ptv_datasets.py [0:0]
118 lines of code
9 McCabe index (conditional complexity)

def Ptvssv2(cfg, mode):
    """
    Construct PyTorchVideo Something-Something v2 SSv2 video loader.
    Load SSv2 data (frame paths, labels, etc. ) to SSv2 Dataset object.
    The dataset could be downloaded from Chrades official website
    (https://20bn.com/datasets/something-something).
    Please see datasets/DATASET.md for more information about the data format.
    For training and validation, a single  clip is randomly sampled from every
    video with random cropping and scaling. For testing, multiple clips are
    uniformaly sampled from every video with uniform cropping. For uniform cropping,
    we take the left, center, and right crop if the width is larger than height,
    or take top, center, and bottom crop if the height is larger than the width.
    Args:
        cfg (CfgNode): configs.
        mode (string): Options includes `train`, `val`, or `test` mode.
    """
    # Only support train, val, and test mode.
    assert mode in [
        "train",
        "val",
        "test",
    ], "Split '{}' not supported".format(mode)

    logger.info("Constructing Ptvcharades {}...".format(mode))

    if mode in ["train", "val"]:
        num_clips = 1
        num_crops = 1

        transform = Compose(
            [
                ApplyTransformToKey(
                    key="video",
                    transform=Compose(
                        [
                            Lambda(div255),
                            NormalizeVideo(cfg.DATA.MEAN, cfg.DATA.STD),
                            RandomShortSideScale(
                                min_size=cfg.DATA.TRAIN_JITTER_SCALES[0],
                                max_size=cfg.DATA.TRAIN_JITTER_SCALES[1],
                            ),
                            RandomCropVideo(cfg.DATA.TRAIN_CROP_SIZE),
                            Lambda(rgb2bgr),
                        ]
                        + (
                            [RandomHorizontalFlipVideo(p=0.5)]
                            if cfg.DATA.RANDOM_FLIP
                            else []
                        )
                        + [PackPathway(cfg)]
                    ),
                ),
                DictToTuple(num_clips, num_crops),
            ]
        )
        clip_sampler = make_clip_sampler(
            "constant_clips_per_video",
            1,  # Put arbitrary duration as ssv2 always needs full video clip.
            num_clips,
            num_crops,
        )
        if cfg.NUM_GPUS > 1:
            video_sampler = DistributedSampler
        else:
            video_sampler = (
                RandomSampler if mode == "train" else SequentialSampler
            )
    else:
        assert cfg.TEST.NUM_ENSEMBLE_VIEWS == 1
        num_clips = cfg.TEST.NUM_ENSEMBLE_VIEWS
        num_crops = cfg.TEST.NUM_SPATIAL_CROPS

        transform = Compose(
            [
                ApplyTransformToKey(
                    key="video",
                    transform=Compose(
                        [
                            Lambda(div255),
                            NormalizeVideo(cfg.DATA.MEAN, cfg.DATA.STD),
                            ShortSideScale(size=cfg.DATA.TEST_CROP_SIZE),
                        ]
                    ),
                ),
                UniformCropVideo(size=cfg.DATA.TEST_CROP_SIZE),
                ApplyTransformToKey(
                    key="video",
                    transform=Compose(
                        [Lambda(rgb2bgr), PackPathway(cfg)],
                    ),
                ),
                DictToTuple(num_clips, num_crops),
            ]
        )
        clip_sampler = make_clip_sampler(
            "constant_clips_per_video",
            1,  # Put arbitrary duration as ssv2 always needs full video clip.
            num_clips,
            num_crops,
        )
        video_sampler = (
            DistributedSampler if cfg.NUM_GPUS > 1 else SequentialSampler
        )

    label_name_file = os.path.join(
        cfg.DATA.PATH_TO_DATA_DIR, "something-something-v2-labels.json"
    )
    video_label_file = os.path.join(
        cfg.DATA.PATH_TO_DATA_DIR,
        "something-something-v2-{}.json".format(
            "train" if mode == "train" else "validation"
        ),
    )
    data_path = os.path.join(
        cfg.DATA.PATH_TO_DATA_DIR,
        "{}.csv".format("train" if mode == "train" else "val"),
    )
    dataset = SSv2(
        label_name_file=label_name_file,
        video_label_file=video_label_file,
        video_path_label_file=data_path,
        clip_sampler=clip_sampler,
        video_sampler=video_sampler,
        transform=transform,
        video_path_prefix=cfg.DATA.PATH_PREFIX,
        frames_per_clip=cfg.DATA.NUM_FRAMES,
        rand_sample_frames=mode == "train",
    )

    logger.info(
        "Constructing ssv2 dataloader (size: {}) from {}".format(
            len(dataset._path_to_videos), data_path
        )
    )

    return PTVDatasetWrapper(
        num_videos=len(dataset._path_to_videos),
        clips_per_video=num_clips,
        crops_per_clip=num_crops,
        dataset=dataset,
    )