+
jh t ^ RI Ht ^ RIt^ RIHt ^ RIHtHt ^RIH t ^RI
H t H t ^RI
Ht ! R R ]4 tR# )
) annotationsN)Path)AnyCallable)default_loader) download_and_extract_archiveverify_str_arg)
VisionDatasetc v a ] tR t^
tRtRtRRRRR]3R V 3R llltR R
ltR R lt R
R lt
R R lt Rt V ;t
# ) FGVCAircrafta `FGVC Aircraft `_ Dataset.
The dataset contains 10,000 images of aircraft, with 100 images for each of 100
different aircraft model variants, most of which are airplanes.
Aircraft models are organized in a three-levels hierarchy. The three levels, from
finer to coarser, are:
- ``variant``, e.g. Boeing 737-700. A variant collapses all the models that are visually
indistinguishable into one class. The dataset comprises 100 different variants.
- ``family``, e.g. Boeing 737. The dataset comprises 70 different families.
- ``manufacturer``, e.g. Boeing. The dataset comprises 30 different manufacturers.
Args:
root (str or ``pathlib.Path``): Root directory of the FGVC Aircraft dataset.
split (string, optional): The dataset split, supports ``train``, ``val``,
``trainval`` and ``test``.
annotation_level (str, optional): The annotation level, supports ``variant``,
``family`` and ``manufacturer``.
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.
download (bool, optional): If True, downloads the dataset from the internet and
puts it in root directory. If dataset is already downloaded, it is not
downloaded again.
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.
zWhttps://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/archives/fgvc-aircraft-2013b.tar.gztrainvalvariantNFc 8 V ^8 d QhRRRRRRRRRRR R
R R R
R/# ) rootz
str | Pathsplitstrannotation_level transformzCallable | Nonetarget_transformdownloadboolloaderzCallable[[str], Any]returnNone )formats "c/home/luca/screen-checking/.venv/lib/python3.14/site-packages/torchvision/datasets/fgvc_aircraft.py __annotate__FGVCAircraft.__annotate__. sZ 0 0 0 0 0
# 0 *
0 0 %0
0 c
< \ SV ` WVR 7 \ VRR4 V n \ VRR4 V n \
P P V P R4 V n V' d V P 4 V P 4 ' g \ R4 h\
P P V P RRR RR
RR /V P , 4 p\ V4 ;_uu_ 4 p V U
u. uF qP 4 NK up
V n R R R 4 \! \# V P \% \' V P 4 4 4 4 V n \
P P V P RR
4 p \
P P V P RRV P RV P R24 p . V n . V n \ V 4 ;_uu_ 4 p V F p
V
P 4 P/ R^4 w rV P* P1 \
P P W R24 4 V P, P1 V P( V, 4 K R R R 4 Wpn R # u up
i + ' g i EL; i + ' g i L1; i))r r r r r family manufacturerzfgvc-aircraft-2013bz;Dataset not found. You can use download=True to download itdataz variants.txtz families.txtzmanufacturers.txtNimagesimages__z.txt z.jpg)trainvalr
test)r r# r$ )super__init__r _split_annotation_levelospathjoinr
_data_path _download
_check_exists RuntimeErroropenstripclassesdictziprangelen class_to_idx _image_files_labelsr appendr )selfr r r r r r r annotation_fileflineimage_data_folder labels_file
image_name
label_name __class__s &&&&&&&& r r. FGVCAircraft.__init__. s EUV$UG5YZ !/ 02W"
'',,tyy2GH NN !!## \] ] '',, OO >. 3
$$
&
/
"
"a567QTJJLQ7DL # !T\\5T\\9J3K!L M GGLL &( Kggll4??FgdF\F\E]]^_c_j_j^kko r@ rC s &r __len__FGVCAircraft.__len__` s 4$$%%r! c V ^8 d QhRRRR/# )r idxrN r ztuple[Any, Any]r )r s "r r r c s
s
r! c V P V, V P V, r2V P V4 pV P ' d V P V4 pV P ' d V P V4 pWC3# rP )r@ rA r r r )rC rU
image_filelabelimages && r __getitem__FGVCAircraft.__getitem__c sf - -c 2DLL4EE J' >>>NN5)E ))%0E|r! c V ^8 d QhRR/# )r r r r )r s "r r r o s ; ;4 ;r! c
v V P 4 ' d R# \ V P V P 4 R# )zG
Download the FGVC Aircraft dataset archive and extract it under root.
N)r6 r _URLr rQ s &r r5 FGVCAircraft._downloado s( $TYY :r! c V ^8 d QhRR/# )r r r r )r s "r r r w s R Rt Rr! c \ P P V P 4 ;' d* \ P P V P 4 # rP )r1 r2 existsr4 isdirrQ s &r r6 FGVCAircraft._check_existsw s2 ww~~doo.QQ277== 3QQr! )r0 r4 r@ rA r/ r? r: r )__name__
__module__ __qualname____firstlineno____doc__r^ r r. rR rZ r5 r6 __static_attributes__
__classcell__)rK s @r r r
sJ < eD
)%),0 '50 0 d&
;R Rr! r )
__future__r r1 pathlibr typingr r folderr utilsr r visionr
r r r! r rr s* " " ? !kR= kRr!