IterableDataset should be added into dataset.pyi

[Rb-Liu]

🐛 Bug

trying to import IterableDataset in torch.utils.data.dataset creates an error in PyCharm. look at the following picture:

It says that cannot find reference in dataset.pyi, where in dataset.pyi, we can see that:

It's obviously that there is no IterableDataset but only its super class Dataset exists.
I believe it should be added.

To Reproduce

Steps to reproduce the behavior:

1. Open PyCharm
2. Write from torch.utils.data.dataset import IterableDataset
3. Error

Expected behavior

No unresolved reference hinted.

Environment

PyTorch version: 1.2.0
Is debug build: No
CUDA used to build PyTorch: Could not collect

OS: Arch Linux
GCC version: (GCC) 9.2.0
CMake version: version 3.15.4

Python version: 3.7
Is CUDA available: No
CUDA runtime version: 10.1.243
GPU models and configuration: GPU 0: GeForce RTX 2060
Nvidia driver version: 435.21
cuDNN version: /usr/lib/libcudnn.so.7.6.4

Versions of relevant libraries:
[pip3] numpy==1.17.2
[pip3] torch==1.2.0
[pip3] torchvision==0.4.0a0
[conda] Could not collect

Additional context

cc @ssnl @ezyang

[ssnl]

Thanks for reporting! Would you mind submitting a PR to fix this?

[DuckSoft]

Pull request submitted! Would you mind a look at #27966?
@ssnl

[Legend94rz]

Still have this problem.
from .dataset import IterableDataset as IterableDataset also need to be added into torch\utils\data\__init__.pyi

[DuckSoft]

ACK. I'll try opening up another pull request to solve this.

[DuckSoft]

Pull request opened. Awaiting to be merged.
cc @ssnl @ezyang