o ϯi @sddlmZddlZddlZddlZddlmZmZmZm Z m Z m Z m Z ddl Z ddlZ ddlmZddlmZddlmZeeZGdd d Ze jjjdd dd d d fd ddZdS)!) annotationsN)AnyCallableListOptionalTupleTypeUnion) g_pathmgr) ClipSampler)LabeledVideoDatasetc@s|eZdZdZgdZddddZe d#d$ddZe d%d&ddZ d'd(ddZ ddZ e de Z d)dd Z d*d!d"Z dS)+Hmdb51LabeledVideoPathsa Pre-processor for Hmbd51 dataset mentioned here - https://serre-lab.clps.brown.edu/resource/hmdb-a-large-human-motion-database/ This dataset consists of classwise folds with each class consisting of 3 folds (splits). The videos directory is of the format, video_dir_path/class_x/.avi ... video_dir_path/class_y/.avi The splits/fold directory is of the format, folds_dir_path/class_x_test_split_1.txt folds_dir_path/class_x_test_split_2.txt folds_dir_path/class_x_test_split_3.txt ... folds_dir_path/class_y_test_split_1.txt folds_dir_path/class_y_test_split_2.txt folds_dir_path/class_y_test_split_3.txt And each text file in the splits directory class_x_test_split_<1 or 2 or 3>.txt <0 or 1 or 2> where 0,1,2 corresponds to unused, train split respectively. Each video has name of format ______.avi For more details on tags - https://serre-lab.clps.brown.edu/resource/hmdb-a-large-human-motion-database/ )r r rr)traintestunusedr data_pathstrsplit_idint split_typereturncszt|}|st|dSt||jvr"t|d|jdSdtt|tfdd|D}| ||S)a6 Factory function that creates Hmdb51LabeledVideoPaths object form a splits/folds directory. Args: data_path (str): The path to the splits/folds directory of HMDB51. split_id (int): Fold id to be loaded. Belongs to [1,2,3] split_type (str): Split/Fold type to be loaded. It belongs to one of the following, - "train" - "test" - "unused" (This is a small set of videos that are neither of part of test or train fold.) z! not found or is not a directory.z! not found in allowed split id's . _test_splitc3s0|]}|r|jdkr|jvr|VqdS).txtN)is_filesuffixstem).0ffile_name_formatL/home/ubuntu/.local/lib/python3.10/site-packages/pytorchvideo/data/hmdb51.py Ssz3Hmdb51LabeledVideoPaths.from_dir..) pathlibPathis_dir RuntimeErrorr_allowed_splitsrsortediterdir from_csvs)clsrrr file_pathsr$r"r%from_dir8s   z Hmdb51LabeledVideoPaths.from_dirr0List[Union[pathlib.Path, str]]c Cs6g}|D]}t|}t|sJ|d|jdkr!d|jvs*td|Sd}||jddd}t |dA}| D]2}| dd }t |d |j|ksZqFtj||d }|d dd d } |||| d fqFWdn1swYqt|d ksJd|d||S)a Factory function that creates Hmdb51LabeledVideoPaths object form a list of split files of .txt type Args: file_paths (List[Union[pathlib.Path, str]]) : The path to the splits/folds directory of HMDB51. split_type (str): Split/Fold type to be loaded. - "train" - "test" - "unused" z not found.rrz Ivalid file: _Nrr ri)label meta_tagszFailed to load dataset from r)r'r(r existsrrr*joinsplitopenread splitlinesrsplitr_split_type_dictospathappendlen) r/r0rvideo_paths_and_label file_path action_namer! path_label line_splitr8r$r$r%r.[s2    z!Hmdb51LabeledVideoPaths.from_csvspaths_and_labels List[Tuple[str, Optional[dict]]]NonecCs||_||_dS)z Args: paths_and_labels [(str, int)]: a list of tuples containing the video path and integer label. N)_paths_and_labels _path_prefix)selfrK path_prefixr$r$r%__init__s z Hmdb51LabeledVideoPaths.__init__cCs ||_dS)N)rO)rPprefixr$r$r%rQs z#Hmdb51LabeledVideoPaths.path_prefixNindexTuple[str, dict]cCs"|j|\}}tj|j||fS)z Args: index (int): the path and label index. Returns: The path and label tuple for the given index. )rNrArBr:rO)rPrTrBr7r$r$r% __getitem__sz#Hmdb51LabeledVideoPaths.__getitem__cCs t|jS)zQ Returns: The number of video paths and label pairs. )rDrN)rPr$r$r%__len__s zHmdb51LabeledVideoPaths.__len__)r r)rrrrrrrr)r)r0r2rrrr)rJ)rKrLrrM)rTrrrU)rr)__name__ __module__ __qualname____doc__r+r@ classmethodr1r.rRrQpropertyrVrWr$r$r$r%rs  "+    rrJrTpyavr pathlib.Path clip_samplerr video_samplerType[torch.utils.data.Sampler] transformOptional[Callable[[dict], Any]]video_path_prefixrrrrdecoderrr c Cs:tjdtj|||d} || _t| |||||d} | S)a A helper function to create ``LabeledVideoDataset`` object for HMDB51 dataset Args: data_path (pathlib.Path): Path to the data. The path type defines how the data should be read: * For a file path, the file is read and each line is parsed into a video path and label. * For a directory, the directory structure defines the classes (i.e. each subdirectory is a class). clip_sampler (ClipSampler): Defines how clips should be sampled from each video. See the clip sampling documentation for more information. video_sampler (Type[torch.utils.data.Sampler]): Sampler for the internal video container. This defines the order videos are decoded and, if necessary, the distributed split. transform (Callable): This callable is evaluated on the clip output before the clip is returned. It can be used for user defined preprocessing and augmentations to the clips. See the ``LabeledVideoDataset`` class for clip output format. video_path_prefix (str): Path to root directory with the videos that are loaded in LabeledVideoDataset. All the video paths before loading are prefixed with this path. split_id (int): Fold id to be loaded. Options are 1, 2 or 3 split_type (str): Split/Fold type to be loaded. Options are ("train", "test" or "unused") decoder (str): Defines which backend should be used to decode videos. zPYTORCHVIDEO.dataset.Hmdb51)rr) decode_audiorf)torch_C_log_api_usage_oncerr1rQr ) rr`rarcrerrrgrflabeled_video_pathsdatasetr$r$r%Hmdb51s / rm)rr_r`r rarbrcrdrerrrrrrfrrr ) __future__rloggingrAr'typingrrrrrrr rhtorch.utils.dataiopath.common.file_ior clip_samplingr labeled_video_datasetr getLoggerrXloggerrutilsdata RandomSamplerrmr$r$r$r%s* $