-
### 🐛 Describe the bug
Minimal example:
```
import os
os.environ["TORCH_LOGS"] = "guards"
os.environ["TORCH_TRACE"] = "."
import torch
import numpy as np
@torch._dynamo.optimize()
def…
rec updated
54 minutes ago
-
Hello
The purpose of this issue to add an sdl/container for running benchmarks for pytorch with [torchbench](https://github.com/pytorch/benchmark/tree/main) on gpu providers.
I have written an…
-
### 🐛 Describe the bug
For con2d's input channel is zero, the inductor gets the wrong size and stride with the eager mod(the eager mod's size and stride are ```(3, 0, 7, 14)``` and
```(98, 98, 14,…
-
Hi,
I am trying to feed as argument image size=300 but I'm getting this error (my images are indeed of size 300)
assert h == img_size and w == img_size, f'height and width of image must be {img_size…
-
### 🐛 Describe the bug
Our team works with diffusion models which utilize a UNet architecture. I am able to successfully compile the model for inference on A100s and get a 50% speed up. But I am runn…
-
### 🐛 Describe the bug
Inductor default
- [x] Dalle2_Pytorch https://github.com/pytorch/pytorch/issues/104797 - @jon-chuang
- [x] basic_gnn_gcn/basic_gnn_sage/basic_gnn_gin : @oulgen
```2023…
-
There's small nuances in how the dynamo runners benchmark models that can make certain torchbench models fail
Some models might be explicitly skipped, others might fail because of some dtype conve…
-
### 🐛 Describe the bug
TorchDynamo does not trace the graph when using custom APEX LayerNorm and this results in no fusion of other elementwise operations. The issue can be reproduced using the code …
-
### 🐛 Describe the bug
The following testcase can reproduce this issue:
```
import torch
def func(x):
# return torch.add(x, 1).numpy()
return torch.Tensor.numpy(torch.add(x, 1))
x =…
-
In [`62bcd02`](https://github.com/prateekralhan/status/commit/62bcd02b4db612389736806306362859762477b8
), ChatGPT + DALL-E (https://streamlit-chatgpt-dalle2.onrender.com/) was **down**:
- HTTP code: 0…