-
### 🐛 Describe the bug
Hi,
I am trying to train and evaluate pre-trained Faster R-CNN model with standard coco dataset. I am getting the following error
TypeError: RandomIoUCrop() requires input …
-
### 🐛 Describe the bug
When using `torch.full((n,), v)` to create a tensor with a dynamic value, one receives a `Pending unbacked symbols` error. A simple workaround is to use `torch.ones((n,)) * v…
-
### 🐛 Describe the bug
## Describe the bug
When using `torch.jit.script` in combination with `torch.onnx.export` the resulting `onnx` graph contains dead code and loops which seem entirely redunda…
-
### 🐛 Describe the bug
# Summary
Using `Tensor.all()` on CUDA tensors with `float32` dtype causes an "illegal memory access" exception after upgrading to PyTorch 2.5 in combination with Python 3…
-
### 🐛 Describe the bug
I have a small script to reproduce how a toy model and the following three features lead to an error when combined:
1. torch.compile
2. FSDP1 with cpu offloading
3. PyTorch …
-
### Your current environment
```text
Collecting environment information...
PyTorch version: 2.3.1
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A
OS: Ub…
-
### Your current environment
```
# python collect_env.py
WARNING 11-08 16:53:16 cuda.py:23] You are using a deprecated `pynvml` package. Please install `nvidia-ml-py` instead, and make sure to un…
Bihan updated
2 weeks ago
-
Feel free to use any online tutorial/repository as a backbone. Ultimately, I want you to be able to explain line-by-line _what_ the code is accomplishing. Choose a CNN on a simple dataset, e.g. MNIST …
-
### 🐛 Describe the bug
Is symbolic execution expected to fail for kernels that have fallthrough registrations?
Reproducer with a fallthrough for the mul operator:
```python
import torch
impo…
-
### 🐛 Describe the bug
Ideally, tensor.triu_(1) should fill lower triangular part with 0. However, it fails to do so when the matrix is large. For example:
```
q_len = 100000
causal_mask = torch.f…