starlight-apk/Qwen
1
0
Fork 0
mirror of https://github.com/QwenLM/Qwen.git synced 2026-05-20 16:35:47 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Junyang Lin
2023-10-08 15:32:23 +08:00
committed by GitHub
parent 05577dea69
commit 581512f6b5
1 changed files with 5 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
README.md
View File

@@ -99,11 +99,13 @@ Before running the code, make sure you have setup the environment and installed
pip install -r requirements.txt pip install -r requirements.txt
``` ```
If your device supports fp16 or bf16, we recommend installing [flash-attention](https://github.com/Dao-AILab/flash-attention) for higher efficiency and lower memory usage. (**flash-attention is optional and the project can run normally without installing it**) If your device supports fp16 or bf16, we recommend installing [flash-attention](https://github.com/Dao-AILab/flash-attention) (**we support flash attention 2 now.**) for higher efficiency and lower memory usage. (**flash-attention is optional and the project can run normally without installing it**)
```bash ```bash
git clone -b v1.0.8 https://github.com/Dao-AILab/flash-attention # Previous installation commands. Now Flash Attention 2 is supported.
cd flash-attention && pip install . # git clone -b v1.0.8 https://github.com/Dao-AILab/flash-attention
# cd flash-attention && pip install .
pip install flash-attn --no-build-isolation
# Below are optional. Installing them might be slow. # Below are optional. Installing them might be slow.
# pip install csrc/layer_norm # pip install csrc/layer_norm
# pip install csrc/rotary # pip install csrc/rotary