\h@ ~ S SK r S SKr S SKrS SKJr S SKJrJrJrJ r S SK
J r SSK J
r
Jr SSKJr " S S \5 rg)
N)Sequence)AnyCallableOptionalUnion)Image ) download_and_extract_archiveverify_str_arg)
VisionDatasetc ^ \ rS rSrSrSrSr SS\\\ R 4 S\S\\ \ \4 S\ \
S \ \
S
\ \
S \4U 4S jjjrS
\4S jrS\S
\\\4 4S jrS
\4S jrSS jrSrU =r$ )
OxfordIIITPet
aX `Oxford-IIIT Pet Dataset `_.
Args:
root (str or ``pathlib.Path``): Root directory of the dataset.
split (string, optional): The dataset split, supports ``"trainval"`` (default) or ``"test"``.
target_types (string, sequence of strings, optional): Types of target to use. Can be ``category`` (default) or
``segmentation``. Can also be a list to output a tuple with all specified target types. The types represent:
- ``category`` (int): Label for one of the 37 pet categories.
- ``binary-category`` (int): Binary label for cat or dog.
- ``segmentation`` (PIL image): Segmentation trimap of the image.
If empty, ``None`` will be returned as target.
transform (callable, optional): A function/transform that takes in a PIL image and returns a transformed
version. E.g, ``transforms.RandomCrop``.
target_transform (callable, optional): A function/transform that takes in the target and transforms it.
transforms (callable, optional): A function/transform that takes input sample
and its target as entry and returns a transformed version.
download (bool, optional): If True, downloads the dataset from the internet and puts it into
``root/oxford-iiit-pet``. If dataset is already downloaded, it is not downloaded again.
))z=https://www.robots.ox.ac.uk/~vgg/data/pets/data/images.tar.gz 5c4f3ee8e5d25df40f4fd59a7f44e54c)zBhttps://www.robots.ox.ac.uk/~vgg/data/pets/data/annotations.tar.gz 95a8c909bbe2e81eed6a22bccdf3f68f)categorybinary-category segmentationrootsplit target_types
transforms transformtarget_transformdownloadc
> [ USS5 U l [ U[ 5 ( a U/nU Vs/ s H n[ USU R 5 PM snU l [ TU ] XXVS9 [ R " U R 5 S- U l U R S- U l U R S- U l
U R S- U l U( a U R 5 U R! 5 ( d [# S 5 e/ n / U l / U l [) U R U R S
3- 5 n
U
H n U R+ 5 R- 5 u ppU R/ U 5 U R$ R/ [1 U
5 S -
5 U R& R/ [1 U5 S -
5 M S S S 5 S S
/U l [5 [7 XR$ 5 V V
s1 s H u pU R9 SS 5 S U
4iM sn
n S S9 VVs/ s H- u nnSR; S UR- S5 5 5 PM/ snnU l [? [7 U R2 [A [C U R2 5 5 5 5 U l" [? [7 U R< [A [C U R< 5 5 5 5 U l# U V s/ s H oR U S3- PM sn U l$ U V s/ s H oR U S3- PM sn U l% g s snf ! , ( d f GNt= fs sn
n f s snnf s sn f s sn f )Nr )trainvaltestr )r r r zoxford-iiit-petimages annotationstrimapsz;Dataset not found. You can use download=True to download itz.txtr CatDog_r c U S $ )Nr )image_id_and_labels ^/var/www/fran/franai/venv/lib/python3.13/site-packages/torchvision/datasets/oxford_iiit_pet.py(OxfordIIITPet.__init__..W s
/A!/D )key c 3 @ # U H oR 5 v M g 7fN)title).0parts r( )OxfordIIITPet.__init__..T s A.@dZZ\\.@s z.jpgz.png)&r _split
isinstancestr_VALID_TARGET_TYPES
_target_typessuper__init__pathlibPathr _base_folder_images_folder _anns_folder _segs_folder _download
_check_exists RuntimeError_labels _bin_labelsopenstripr appendint bin_classessortedziprsplitjoinclassesdictrangelenbin_class_to_idx class_to_idx_images_segs)selfr r r r r r r target_type image_idsfilelineimage_idlabel bin_labelr$ raw_cls __class__s r( r; OxfordIIITPet.__init__+ s %UG5IJ lC ( ( (>Leq
eqVaN;8P8P Qeq
m #LL 36G G " / /( : - -
= - - 9 NN !!## \] ]
$##
T&::
;t 04
0B0B0D- * ##CJN3 ''I(:; < "5> %LOPY[g[gLhiLh (//#q)!,e4LhiD
HHAgmmC.@A A
!%S)9)95TEUEUAV;W%X Y T\\5T\\9J3K!L MPYZPYH++
$.??PYZ LUVI''XJd*;;IV
M
& <
; j
[Vs* !L)7B
L./#M
4M M M.
L=returnc , [ U R 5 $ r/ )rS rV )rX s r( __len__OxfordIIITPet.__len__` s 4<< r+ idxc . [ R " U R U 5 R S5 n/ nU R H nUS:X a UR U R U 5 M) US:X a UR U R U 5 MO UR [ R " U R U 5 5 M U( d S nO [ U5 S:X a US nO [ U5 nU R ( a U R X#5 u p#X#4$ )NRGBr r r r ) r rG rV convertr9 rI rE rF rW rS tupler )rX rg imagetargetrY s r( __getitem__OxfordIIITPet.__getitem__c s
4<<,-55e< --Kj(
dll3 /0 11
d . .s 34
ejjC 9:
. F
[A
AYF6]F ?? OOE :ME} r+ c U R U R 4 HM n[ R R U5 ( a& [ R R U5 ( a MM g g)NFT)r? r@ ospathexistsisdir)rX folders r( rC OxfordIIITPet._check_exists{ sJ **D,=,=>FGGNN6**rww}}V/D/D ? r+ c U R 5 ( a g U R H# u p[ U[ U R 5 US9 M% g )N)
download_rootmd5)rC
_RESOURCESr
r7 r> )rX urlry s r( rB OxfordIIITPet._download s< HC (C@Q@Q rF rV r? rE rW rA r5 r9 rT rK rU rP )r r NNNF)rc N)__name__
__module__ __qualname____firstlineno____doc__rz r8 r r7 r< r= r r r boolr; rJ re rk r rn rC rB __static_attributes__
__classcell__)ra s @r( r r
s .J J
2<)-(,/3 3WC %&3W 3W HSM3 ./ 3W
X& 3W H%
3W #8,3W 3W 3Wj! ! s uS#X 0t ] ]r+ r )rq os.pathr< collections.abcr typingr r r r PILr utilsr
r visionr r r& r+ r( r s0 $ 1 1 ? !z]M z]r+