-
### Your current environment
```
Collecting environment information...
PyTorch version: 2.0.1+cu117
Is debug build: False
CUDA used to build PyTorch: 11.7
ROCM used to build PyTorch: N/A
OS…
-
# Prerequisites
- [x] I am running the latest code. Development is very rapid so there are no tagged versions as of now.
- [x] I carefully followed the [README.md](https://github.com/abetlen/llama-c…
-
### Your current environment
```text
Collecting environment information...
PyTorch version: 2.1.2+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A
…
-
### What version of Bun is running?
0.3.0
### What platform is your computer?
Linux 5.15.49-linuxkit #1 SMP Tue Sep 13 07:51:46 UTC 2022 x86_64 unknown
### What steps can reproduce the bug…
yvz5 updated
9 months ago
-
### 🐛 Describe the bug
There is a system memory leak when using different input sizes to `torch.nn.Conv3d` on the GPU. A very simple script to reproduce the issue is:
```python
import gc
impor…
-
### 🐛 Describe the bug
I am using torch.distributed.all_to_all_single function do alltoall communication. My input and output tensor are both 2d tensor. From my error code, it seems this function onl…
-
We are running IMB-MPI1 on Broadcom's 100G adaptor on a 2 nodes setup connected back to back and configured a basi QOS scheme between the two nodes. The hosts has been installed with OpenMPI-4.0.4 wit…
-
Evaluating the jvp of a model's forward method with respect to an input variable fails when the model is parallel distributed with DistributedDataParallel. Specifically, the next jvp evaluation after …
-
### Your current environment
```text
Collecting environment information...
PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A
…
-
### 🐛 Describe the bug
Test code:
```python
import torch
base = torch.randn([1,1,1,1,1])
temp_input = torch.quantize_per_tensor(base, 0.1, 10, torch.quint2x4)
module = torch.nn.Upsample(scal…