+
j@ ^ RI t ^ RIt^ RIHt ^ RIHtHtHtHt ^ RI H
t
^RI H t H
t
^RIHt ^RIHt ! R R ]4 tR# )
N)Path)AnyCallableOptionalUnion)Tensor) find_classes make_dataset)
VideoClips)
VisionDatasetc a a ] tR t^
t oRtRtRRRR/t^t^tRV3R lV 3R lllt ]
V3R lR
l4 t V3R lR lt V3R
lR lt
V3R lR ltRtVtV ;t# )HMDB51a>
`HMDB51 `_
dataset.
HMDB51 is an action recognition video dataset.
This dataset consider every video as a collection of video clips of fixed size, specified
by ``frames_per_clip``, where the step in frames between each clip is given by
``step_between_clips``.
To give an example, for 2 videos with 10 and 15 frames respectively, if ``frames_per_clip=5``
and ``step_between_clips=5``, the dataset size will be (2 + 3) = 5, where the first two
elements will come from video 1, and the next three elements from video 2.
Note that we drop clips which do not have exactly ``frames_per_clip`` elements, so not all
frames in a video might be present.
Internally, it uses a VideoClips object to handle clip creation.
Args:
root (str or ``pathlib.Path``): Root directory of the HMDB51 Dataset.
annotation_path (str): Path to the folder containing the split files.
frames_per_clip (int): Number of frames in a clip.
step_between_clips (int): Number of frames between each clip.
fold (int, optional): Which fold to use. Should be between 1 and 3.
train (bool, optional): If ``True``, creates a dataset from the train split,
otherwise from the ``test`` split.
transform (callable, optional): A function/transform that takes in a TxHxWxC video
and returns a transformed version.
output_format (str, optional): The format of the output video tensors (before transforms).
Can be either "THWC" (default) or "TCHW".
Returns:
tuple: A 3-tuple with the following entries:
- video (Tensor[T, H, W, C] or Tensor[T, C, H, W]): The `T` video frames
- audio(Tensor[K, L]): the audio frames, where `K` is the number of channels
and `L` is the number of points
- label (int): class of the video clip
zJhttps://serre-lab.clps.brown.edu/wp-content/uploads/2013/10/hmdb51_org.rarurlzQhttps://serre-lab.clps.brown.edu/wp-content/uploads/2013/10/test_train_splits.rarmd5 15e67781e70dcfbdce2d7dbb9b3344b5c < V ^8 d QhRS[ S[S[3, RS[RS[RS[RS[S[, RS[RS[RS[S[, R S[S[S[S[3, , R
S[R S[R S[R
S[RS[RS[RR/# ) rootannotation_pathframes_per_clipstep_between_clips
frame_ratefoldtrain transform_precomputed_metadata num_workers _video_width
_video_height_video_min_dimension_audio_samples
output_formatreturnN) r strr intr boolr dictr )format
__classdict__s "\/home/luca/screen-checking/.venv/lib/python3.14/site-packages/torchvision/datasets/hmdb51.py __annotate__HMDB51.__annotate__= s 4# 4#CI 4# 4# 4#
4# SM
4# 4# 4# H%4# (S#X 74# 4# 4# 4# " 4# 4# !4#"
#4# c < \ SV ` V4 VR9 d \ RV 24 hRp\ V P 4 w V n p\
V P VV4 V n V P UUu. uF w ppVNK
ppp\ VVVVV V
V V V
VVR7 pVV n W`n
Wpn V P VW&V4 V n
VP V P 4 V n Wn R# u uppi ) z$fold should be between 1 and 3, got )r r r r r! r" N)r/ r )avi)super__init__
ValueErrorr r classesr
samplesr full_video_clipsr r _select_foldindicessubset video_clipsr )selfr r r r r r r r r r r r r r! r"
extensions class_to_idxpath_ video_pathsr; __class__s &&&&&&&&&&&&&&&& r* r3 HMDB51.__init__= s $ y CD6 JK K
%1$))%<" l# II
.2\\:\ qt\ : !#%'!5)'
!,
((oUS &--dll;"- ;s 0C*c 6 < V ^8 d QhRS[ S[S[3, /# r r# )r' r$ r )r( r) s "r* r+ r, t s . .$sCx. .r- c . V P P # N)r7 metadatar< s &r* rH HMDB51.metadatas s $$---r- c X < V ^8 d QhRS[ S[, RS[RS[RS[RS[ S[, /# )r
video_listannotations_dirr r r# )listr$ r% r& )r( r) s "r* r+ r, w s< tCy 3 c Z^ cghkcl r- c | V' d
V P M V P pR V R2p\ P P W&4 p\
P
! V4 p\
4 p V Fo p
\ V
4 ;_uu_ 4 p V P 4 p RRR4 X F: p
V
P 4 w r\ V4 pVV8X g K) V P V4 K< Kq . p\ V4 F= w pp\ P P V4 V 9 g K, VP V4 K? V# + ' g i L; i)z *test_splitz.txtN) TRAIN_TAGTEST_TAGosr? joinglobsetopen readlinessplitr% add enumeratebasenameappend)r< rL rM r r
target_tagsplit_pattern_namesplit_pattern_pathannotation_pathsselected_filesfilepathfidlineslinevideo_filename
tag_stringtagr9 video_index
video_paths &&&&& r* r8 HMDB51._select_foldw s ',T^^$--
*4& 5 WW\\/ N 99%78 (Hh 3
-1ZZ\**o*$"&&~6 ) '0'< #Kww
+~={+ (= s >D++ D; c < V ^8 d QhRS[ /# rE )r% )r( r) s "r* r+ r, s , , ,r- c 6 V P P 4 # rG )r; num_clipsrI s &r* __len__HMDB51.__len__ s ))++r- c @ < V ^8 d QhRS[ RS[S[S[S[ 3, /# )r idxr# )r% tupler )r( r) s "r* r+ r, s' ) )s )uVVS-@'A )r- c V P P V4 w r#rEV P V, pV P V, w rGV P e V P V4 pW#V3# rG )r; get_clipr9 r6 r )r< rr videoaudior@ video_idx sample_index class_indexs && r* __getitem__HMDB51.__getitem__ s^ %)%5%5%>%>s%C"a||I. l3 >> %NN5)E[((r- )r5 r r7 r9 r6 r r r; ) r/ Nr/ TNNr/ r r r r THWC)__name__
__module__ __qualname____firstlineno____doc__data_urlsplitsrP rQ r3 propertyrH r8 ro r{ __static_attributes____classdictcell__
__classcell__)rB r) s @@r* r r
st %N \H
b
1F IH4# 4#l . . ,, ,) ) )r- r )rT rR pathlibr typingr r r r torchr folderr r
video_utilsr visionr r r- r* r s. 1 1 . # !K)] K)r-