-
### Your current environment
```text
PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A
OS: Ubuntu 22.04.4 LTS (x86_64)
GCC …
-
### 🐛 Describe the bug
I am using something very similar to [`transformers.utils.ModelOutput`](https://github.com/huggingface/transformers/blob/d533465150532b0c5de167b574e59f64c68b1154/src/transfor…
-
Hey @lezwon, hey @yassineAlouini. We (`torchvision` maintainers) picked up on the fact that you are very eager to contribute. We are also aware contributing is not a very good experience right now due…
-
### 🐛 Describe the bug
Training throughput slow down when loaded optimizer shard state dict.
1) train model and save model dict/optimizer dict with sharded state, the training throughput like:
,…
-
### 🐛 Describe the bug
AMP static shape default wrapper performance regression
suite
name
thread
batch_size_new
speed_up_new
inductor_new
…
-
This RFC was re-created due to a problem with the original. Summary of comments from previous issue below.
### 🚀 The feature
**TL;DR \-** We want to lean into **modular Multi-Threading/Multi-Pr…
-
### Your current environment
pip3 install vllm==0.4.2 nvidia-ammo==0.7.1
Collecting environment information...
PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used to build PyTorch: …
-
### 🐛 Describe the bug
I'm experiencing a torch.fft.irfft shape mismatch when exporting a model using torch.onnx.dynamo_export. Details below
### Error logs
I've got shapes for repro below
`…
-
### 🐛 Describe the bug
```py
t = torch.tensor([1,2,3,3,4,5,6,7,8], dtype=torch.float)
torch.histc(t, 5, 7, 7)
```
The output given by above code:
```sh
tensor([2., 2., 2., 1., 2.])
```
Acc…