TYP: declare torch.utils in __all__

[twoertwein]

TYP: declare torch.utils in __all__

[kalium222]

same for me.

Version

PyTorch version: 2.3.0+cu118
Is debug build: False
CUDA used to build PyTorch: 11.8
ROCM used to build PyTorch: N/A

OS: Arch Linux (x86_64)
GCC version: (GCC) 14.1.1 20240507
Clang version: 17.0.6
CMake version: version 3.28.1
Libc version: glibc-2.39

Python version: 3.12.3 (main, Apr 23 2024, 09:16:07) [GCC 13.2.1 20240417] (64-bit runtime)
Python platform: Linux-6.9.1-zen1-1-zen-x86_64-with-glibc2.39
Is CUDA available: True
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: GPU 0: NVIDIA GeForce RTX 4070 SUPER
Nvidia driver version: 550.78
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 39 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 8
On-line CPU(s) list: 0-7
Vendor ID: GenuineIntel
Model name: Intel(R) Core(TM) i7-1065G7 CPU @ 1.30GHz
...

Versions of relevant libraries:
[pip3] numpy==1.26.4
[pip3] torch==2.3.0+cu118
[pip3] torchaudio==2.3.0+cu118
[pip3] torchvision==0.18.0+cu118
[conda] Could not collect

[albanD]

Does submodules need to be in the __all__ in general?

[twoertwein]

Does submodules need to be in the __all__ in general?

Not when importing them using from torch import utils or import torch.utils.

The torch namespace does not directly import utils, it seems to be a side effect of from ._tensor import Tensor which makes utils available through the torch namespace.

I'm now using import torch.utils.data so that torch.utils.data.Dataset and torch.utils.data.Dataloader are understood by type checkers.

Feel free to close this issue.

[XuehaiPan]

I opened a PR to resolve this.

• #127703