-
### 🐛 Describe the bug
torch.autograd.profiler.profile triggered a crash when the gpu is available.
minimal example:
```
import threading
import torch
from torch.autograd.profiler import pro…
-
- Site: [http://localhost:8082](http://localhost:8082)
**New Alerts**
- **Insufficient Site Isolation Against Spectre Vulnerability** [90004] total: 1:
- [http://localhost:8082/bookings](htt…
-
### 🐛 Describe the bug
torch.argmax is slow on CPU relative to the numpy implementation. I would expect these two to be approximately equally fast. Here is a small script to reproduce this:
```pyt…
-
### 🐛 Describe the bug
''' update synthetic data '''
if 'BN' not in args.dlmodel: # for ConvNet
loss = torch.tensor(0.0).to(args.dldevice)
for c in …
-
### 🐛 Describe the bug
import numpy as np
import torch
data = torch.tensor(np.array([[1, 2], [3, 4], [5, 6]], dtype=np.float32))
q1 = torch.quantile(data, 0.25, dim=0) # works
data = torch.tens…
-
### 🐛 Describe the bug
If I use Inductor to compile `Conv2d` and `BN2d` **not use** `eval mode`. The results are inconsistent.
The problem seems to be with BN2d
```python
import torch
import …
-
### Your current environment
The output of `python collect_env.py`
```text
Collecting environment information...
WARNING 08-22 15:09:07 _custom_ops.py:14] Failed to import from vllm._C with Mo…
-
### Your current environment
2024-04-24 06:04:07 (27.2 MB/s) - ‘collect_env.py’ saved [24877/24877]
Collecting environment information...
PyTorch version: 2.2.1+cu121
Is debug build: False
CUDA…
-
### 🐛 Describe the bug
Repro:
```python
def test_cond_symint_input(self):
class M(torch.nn.Module):
def forward(self, x, y, z):
a = y.shape[0]
…
-
### 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 importin…