-
### 🚀 The feature, motivation and pitch
I use torch 2.3.1 on MacOS 15.1 beta
I profiling my model and observed that aten::bernoulli is taking 34% of the runtime on cpu while dropout time is marg…
-
# Brief describe
add shape mismatch for this input variation
```
"Tensor self = ?", "Tensor other = 0.0"
```
Run [`pytest tests/pattern/test_add_shape_mismatch.py`](https://github.com/tenstorrent/pyt…
-
Hi :)
i tried training nnUNet with the skeletall recall loss on my dataset und an error has occured. This is listed below. The same dataset was also used on default nnunet itself and there was no …
-
when running examples in **Basic Usage**, it shows errors:
```shell
2024-09-07 20:09:29,033 - modelscope - INFO - PyTorch version 2.0.1 Found.
2024-09-07 20:09:29,033 - modelscope - INFO - Loadin…
-
Canonical ops: https://docs-preview.pytorch.org/90644/ir.html
The list below is an incomplete list of canonical ops + other ops ordered by usage. canonical ops are marked with `canonical`. Be sure…
-
Why do I get an error when using ceDice as the loss function during training: `../aten/src/ATen/native/cuda/NLLLoss2d.cu:104: nll_loss2d_forward_kernel: block: [0,0,0], thread: [278,0,0] Assertion 't …
-
[TL,DR]
The following case is a PyTorch UT reproducer. It has regression with the commit b6cdccd33d901a2c187719e976e0aed026ea23f0 compared with 91b14bf5593cf58a8541f3e6b9125600a867d4ef. It means that…
-
运行第一批次没问题，但后面马上就报错，使用的是源代码在IRSTD-1k上训练的：
RuntimeError: CUDA error: device-side assert triggered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below mi…
-
Consider the following model and export where we use the same tensor as the sample input
```python
import torch
from typing import List
class ListAdd(torch.nn.Module):
def __init__(self…
-
Tracking op support for TorchToLinalg lowering. The ops are ordered based on priority (highest -> lowest) in each section.
**IMPORTANT:** Mark ops you are working on by hovering over it in the list…