-
### 🐛 Describe the bug
**Description:**
I have encountered some precision differences when using the `dropout` and `linear` in the following way.
**Code to Reproduce:**
```python
import tor…
-
### 🐛 Describe the bug
see the log outputs here: https://gist.github.com/alita-moore/a23f57d2adfaa885185308e922aef44a
To reproduce this run the following code:
```
import os
import torch
imp…
-
### 🐛 Describe the bug
Passing this tensor of complex numbers with a large p value to torch.nn.functional.normalize produces nan values on cpu, but eliminating the imaginary part results in 0 in thos…
-
### 🐛 Describe the bug
Under specific inputs, torch.profiler.profile triggered a crash.
minimal example:
```
import torch
import torch.profiler
model = torch.nn.Sequential(
torch.nn.Conv2d(…
x0w3n updated
2 weeks ago
-
### 🐛 Describe the bug
Hi, I found output tensor of `torch.split_with_sizes_copy` has version 0 on CUDA while has version 1 on CPU and XPU.
**Then what's the correct version of output tensor fro…
-
### 🐛 Describe the bug
fused_scaled_matmul_reduce_scatter works for scalar scale but not for channel-wise scale.
A minimal repro: https://gist.github.com/donglinz/9d8cb3ec7f3b6bfb6b4a7d9402c32a…
-
Hello,
I want to apply Latent Diffusion Model to medical images data. How can I feed training images from a directory of .jpg files to the train the diffusion model ?
Plus, I don't want the mode…
-
### 🐛 Describe the bug
torch.cuda.current_stream/default_stream a false INTERNAL ASSERT FAILED when the device does not exist, accompanied by the message: "please report a bug to PyTorch."
**torch…
x0w3n updated
2 weeks ago
-
### Your current environment
The output of `python collect_env.py`
```text
root@vllm-cpu:/workspace# python3 collect_env.py
Collecting environment information...
INFO 08-20 07:37:37 import…
-
### 🐛 Describe the bug
Bug as shown in the title.
Solution: export CUDA_PATH=\~/anaconda3/envs/\:\~/anaconda3/envs/\/targets/x86_64-linux
### Versions
Collecting environment information...
PyTor…