Ei" S SK r S SKrS SKrS SKJr S SKJr S SKJr S SK J
r
J r J r S SK
r
SSKJr SSKJrJrJr S S
S S .rS
r " S S\5 rSS\ \\4 S\ \ S\\\\4 \\ 4 4S jjrS\ \\4 S\S\SS4S jr SS\ \\4 S\ \ SS4S jjr S S\ \\4 S\ \ S\SS4S jjr S S\ \\4 S\ \ S\ \\ S\SS4
S jjr g) N)Iterator)contextmanager)Path)AnyOptionalUnion ) ImageFolder)check_integrityextract_archiveverify_str_arg)zILSVRC2012_img_train.tar 1d675b47d978889d74fa0da5fadfb00e)zILSVRC2012_img_val.tar 29b22e2961454d5413ddabcf34fc5622)z ILSVRC2012_devkit_t12.tar.gz fa75699e90414af021442c21a62c3abf)trainvaldevkitzmeta.binc ~ ^ \ rS rSrSrS
S\\\4 S\S\SS4U 4S jjjr SS jr
\ S\4S
j5 r S\4S jr
S rU =r$ )ImageNet a `ImageNet `_ 2012 Classification Dataset.
.. note::
Before using this class, it is required to download ImageNet 2012 dataset from
`here `_ and
place the files ``ILSVRC2012_devkit_t12.tar.gz`` and ``ILSVRC2012_img_train.tar``
or ``ILSVRC2012_img_val.tar`` based on ``split`` in the root directory.
Args:
root (str or ``pathlib.Path``): Root directory of the ImageNet Dataset.
split (string, optional): The dataset split, supports ``train``, or ``val``.
transform (callable, optional): A function/transform that takes in a PIL image or torch.Tensor, depends on the given loader,
and returns a transformed version. E.g, ``transforms.RandomCrop``
target_transform (callable, optional): A function/transform that takes in the
target and transforms it.
loader (callable, optional): A function to load an image given its path.
By default, it uses PIL as its image loader, but users could also pass in
``torchvision.io.decode_image`` for decoding image data into tensors directly.
Attributes:
classes (list): List of the class name tuples.
class_to_idx (dict): Dict with items (class_name, class_index).
wnids (list): List of the WordNet IDs.
wnid_to_idx (dict): Dict with items (wordnet_id, class_index).
imgs (list): List of (image path, class_index) tuples
targets (list): The class_index value for each image in the dataset
rootsplitkwargsreturnNc > [ R R U5 =ol [ USS5 U l U R
5 [ U R 5 S n[ T U ]$ " U R 40 UD6 Xl U R U l U R U l U R Vs/ s H oTU PM snU l [ U R 5 VVVs0 s H u pgU H oU_M M snnnU l
g s snf s snnnf )Nr )r r r )ospath
expanduserr r
r parse_archivesload_meta_filesuper__init__ split_folderclasseswnids class_to_idx wnid_to_idx enumerate)
selfr r r wnid_to_classeswnididxclsscls __class__s
Z/home/mande/repo/quber/.venv/lib/python3.13/site-packages/torchvision/datasets/imagenet.pyr" ImageNet.__init__4 s 77--d33y#E74DE
(3A6
**5f5 \\
,,:>**E*$-*E 7@ 7N _7N)#Z^SV#XZ^S7N _ F _s 2D!D c [ [ R R U R [
5 5 ( d [
U R 5 [ R R U R 5 ( dM U R S:X a [ U R 5 g U R S:X a [ U R 5 g g g )Nr r ) r r r joinr META_FILEparse_devkit_archiveisdirr# r parse_train_archiveparse_val_archiver) s r0 r ImageNet.parse_archivesC s rww||DIIy ABB +ww}}T . .//zzW$#DII.u$!$)), % 0 c j [ R R U R U R 5 $ N)r r r3 r r r9 s r0 r# ImageNet.split_folderM s ww||DIItzz22r; c : SR " S0 U R D6$ )NzSplit: {split} )format__dict__r9 s r0
extra_reprImageNet.extra_reprQ s &&7 77r; )r& r$ r r r' r% )r )r N)__name__
__module__ __qualname____firstlineno____doc__r strr r r" r propertyr# rC __static_attributes__
__classcell__)r/ s @r0 r r so 8
`U39 -
`c
`s
`W[
`
` - 3c 3 38C 8 8r; r r filer c Uc [ n[ R R X5 n[ U5 ( a [
R " USS9$ Sn[ UR X5 5 e)NT) weights_onlyzThe meta file {} is not present in the root directory or is corrupted. This file is automatically created by the ImageNet dataset.) r4 r r r3 r torchload RuntimeErrorrA )r rN msgs r0 r r U s[ |
77<< #Dt zz$T22
J 3::d122r; md5c [ [ R R X5 U5 ( d Sn[ UR X5 5 eg )Nz{The archive {} is not present in the root directory or is corrupted. You need to download it externally and place it in {}.)r r r r3 rS rA )r rN rU rT s r0 _verify_archiverW d sB 277<<3S 9 9
E 3::d122 :r; c
^ SSK Jm S[ S[ [ [
[ 4 [ [ [ [ S4 4 4 4U4S jjnS[ S[ [
4S jn[ S[ [ 4S j5 n[ S nUc US nUS
n[ XU5 U" 5 n[ [ R R X5 U5 [ R R US 5 nU" U5 u pU" U5 n U V s/ s H oU PM n
n [ R " X4[ R R U [" 5 5 SSS5 gs sn f ! , ( d f g= f) a1 Parse the devkit archive of the ImageNet2012 classification dataset and save
the meta information in a binary file.
Args:
root (str or ``pathlib.Path``): Root directory containing the devkit archive
file (str, optional): Name of devkit archive. Defaults to
'ILSVRC2012_devkit_t12.tar.gz'
r N devkit_rootr .c > [ R R U SS5 nT
R USS9S n[ [ U6 5 S n[
U5 VVs/ s H u pEUS:X d M
X$ PM nnn[ [ U6 5 S S u pgnU V s/ s H n [ U R S 5 5 PM nn [ Xg5 VV
s0 s H u pJXJ_M n nn
[ Xx5 V
V s0 s H u pX_M n n
n X4$ s snnf s sn f s sn
nf s sn n
f )
Ndatazmeta.matT)
squeeze_mesynsets r z, ) r r r3 loadmatlistzipr( tupler )rY metafilemeta
nums_childrenr, num_childrenidcsr% r$ r- r+ idx_to_wnidr* sios r0 parse_meta_mat,parse_devkit_archive..parse_meta_matx s 77<< VZ@{{8{5i@ S$Z(+
3<]3Ka3K /c|_`O` 3Ka #CJ / 3 W7>?wt5D)*w?25d2BC2BYSsy2B C8;E8KL8K*$4:8KL++ b?CLs C5(C5$C;D !Dc [ R R U SS5 n[ U5 nUR 5 nS S S 5 W Vs/ s H n[ U5 PM sn$ ! , ( d f N*= fs snf )Nr[ z&ILSVRC2012_validation_ground_truth.txt)r r r3 open readlinesint)rY rN txtfhval_idcsval_idxs r0 parse_val_groundtruth_txt7parse_devkit_archive..parse_val_groundtruth_txt sV ww||K1YZ
$Z5 (H ,45HG H55 Z5s A"
A3"
A0c 3 # [ R " 5 n U v [ R " U 5 g ! [ R " U 5 f = f7fr= )tempfilemkdtempshutilrmtree)tmp_dirs r0 get_tmp_dir)parse_devkit_archive..get_tmp_dir s5 ""$ #M MM' "FMM' "s A4 AA Ar r ILSVRC2012_devkit_t12)scipy.ioiorJ rc dictrp ra r r ARCHIVE_METArW r r r r3 rQ saver4 )r rN rk rt r| archive_metarU r{ rY ri r* rr r, val_wnidsrj s @r0 r5 r5 m s: ,C ,E$sCx.$sERUWZRZOG[B\2\,] ,6s 6tCy 6 ## # # )L |A
q/CD$ ' T0':ggll7,CD '5k'B$ ,[919:#% :
O/ dI1NO
;
s (A E E# U H% n[ R R TU5 v M' g 7fr= )r r r3 ).0imageval_roots r0 $parse_val_archive.. s% T?SeBGGLL511?Ss -0)r r rW r r r3 r sortedr setmkdirrb ry movebasename)
r rN r% r r rU imagesr+ img_filer s
@r0 r8 r8 s &L |A
q/C } t$Q'D$ww||D)HBGGLL,h7
Trzz(?ST
TFE
h-. e, Hbggll8T277;K;KH;U VW -r; r= )Nr )NNr ) r ry rw collections.abcr
contextlibr pathlibr typingr r r rQ r r
utilsr r r
r r4 r rJ rc r ra r rW r5 r7 r8 r@ r; r0 r sq
$ % ' ' C C N IR
;8{ ;8| 3sDy) 3# 3%PTUXZ]U]P^`deh`iPiJj 3 3%T * 3# 3C 3D 33PuS$Y / 3Px} 3PPT 3PlUeCI . Uhsm UTW Ufj U6 jo!X
T
!X"*3-!X?GS ?R!Xcf!X !Xr;