-
### 🐛 Describe the bug
```
import torch
from torch_geometric.data import Data
from torch_geometric.utils.convert import to_dgl
# 1nd example zero edge
data = Data()
data['x'] = torch.tensor…
-
### 提交前必须检查以下项目
- [X] 请确保使用的是仓库最新代码（git pull）
- [X] 已阅读[项目文档](https://github.com/ymcui/Chinese-LLaMA-Alpaca-3/wiki)和[FAQ章节](https://github.com/ymcui/Chinese-LLaMA-Alpaca-3/wiki/常见问题)并且已在Issue中对问题进…
-
### 🐛 Describe the bug
ffc202a1b91def8c81a6eb9a39777bc7e149e1ee
```
/workspace/pytorch# bash inductor_single_run.sh single inference accuracy huggingface LayoutLMForSequenceClassification amp first…
-
### 🐛 Describe the bug
This issue is a follow-up to https://github.com/pytorch/pytorch/issues/121621 . Basically the same issue, different op.
The op signature from https://pytorch.org/docs/stable/g…
-
### Search before asking
- [X] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and found no similar bug report.
### YOLOv8 Component
_No response_
### Bug
…
mixxs updated
3 weeks ago
-
使用qwen2 7b在业务数据集后sft后，再使用kto训练loss nan，求教可能是什么原因？
deepspeed --num_gpus 8train_kto.py \
--save_path xx \
--save_steps 300 \
--logging_steps 2 \
--micro_train_batch_size 2 \
…
-
### 🐛 Describe the bug
Hi,
I'm trying to make my [MoE Triton kernel](https://github.com/RobertCsordas/moe_layer/blob/master/triton_src/moe_layer/cvmm.py) work with torch.compile(). I know that thi…
-
To allow Clay to run on not just Linux operating systems, but OSX as well!
This comes from user feedback during DevSeed's Team Week, where we found out that the folks developing the Clay model are …
-
**Describe the Bug**
Hey,
I have been trying to install apex in colab but it has the bug as mentioned above in the subject.
It looks like it has something with the amp_C extension.
_Here is the …
-
### 🐛 Describe the bug
Hi,
I found the collective ops in fx graph can be erased by DCE, when the result of collective ops is not used in this rank.
It's not expected, since collective ops have si…