Number of classes in FGVCAircraft differs from documentation

[detkov]

🐛 Describe the bug

In the torchvision.datasets.FGVCAircraft dataset there are 100 classes — not 102, as it said in the documentation and on the official dataset's page.

How to reproduce:

from torchvision.datasets import FGVCAircraft 
ds = FGVCAircraft('./', split='train', download=True)
print(len(set(ds._labels)))

Versions

PyTorch version: 1.11.0+cu102
Is debug build: False
CUDA used to build PyTorch: 10.2
ROCM used to build PyTorch: N/A

OS: Debian GNU/Linux 10 (buster) (x86_64)
GCC version: (Debian 8.3.0-6) 8.3.0
Clang version: Could not collect
CMake version: version 3.13.4
Libc version: glibc-2.10

Python version: 3.7.12 | packaged by conda-forge | (default, Oct 26 2021, 06:08:53)  [GCC 9.4.0] (64-bit runtime)
Python platform: Linux-4.19.0-18-cloud-amd64-x86_64-with-debian-10.12
Is CUDA available: True
CUDA runtime version: 11.0.221
GPU models and configuration: GPU 0: Tesla V100-SXM2-16GB
Nvidia driver version: 460.73.01
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

Versions of relevant libraries:
[pip3] mypy-extensions==0.4.3
[pip3] numpy==1.19.5
[pip3] torch==1.11.0
[pip3] torchaudio==0.10.0+cu113
[pip3] torchvision==0.12.0
[conda] blas                      2.113                       mkl    conda-forge
[conda] blas-devel                3.9.0            13_linux64_mkl    conda-forge
[conda] cudatoolkit               11.1.1              h6406543_10    conda-forge
[conda] dlenv-pytorch-1-10-gpu    1.0.20220226     py37h0ee201a_0    file:///tmp/conda-pkgs
[conda] libblas                   3.9.0            13_linux64_mkl    conda-forge
[conda] libcblas                  3.9.0            13_linux64_mkl    conda-forge
[conda] liblapack                 3.9.0            13_linux64_mkl    conda-forge
[conda] liblapacke                3.9.0            13_linux64_mkl    conda-forge
[conda] mkl                       2022.0.1           h8d4b97c_803    conda-forge
[conda] mkl-devel                 2022.0.1           ha770c72_804    conda-forge
[conda] mkl-include               2022.0.1           h8d4b97c_803    conda-forge
[conda] mypy_extensions           0.4.3            py37h89c1867_4    conda-forge
[conda] numpy                     1.19.5           py37h038b26d_2    conda-forge
[conda] pytorch-mutex             1.0                        cuda    pytorch
[conda] torch                     1.11.0                   pypi_0    pypi
[conda] torchaudio                0.10.0+cu113             pypi_0    pypi
[conda] torchvision               0.12.0                   pypi_0    pypi

cc @pmeier @YosuaMichael

[pmeier]

Huh, thanks for the report. I guess we never bothered to check. Note that the documentation for annotation_level="manufacturer"'s is also wrong:

vision/torchvision/datasets/fgvc_aircraft.py

Lines 20 to 23 in 467b841

	- ``variant``, e.g. Boeing 737-700. A variant collapses all the models that are visually
	indistinguishable into one class. The dataset comprises 102 different variants.
	- ``family``, e.g. Boeing 737. The dataset comprises 70 different families.
	- ``manufacturer``, e.g. Boeing. The dataset comprises 41 different manufacturers.

It should be 30.

Do you want to send a PR?

[detkov]

@pmeier sure, did it: #5814