gnh1201/Open-Llama
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update readme

Browse Source
This commit is contained in:
LiangSong 2023-04-28 22:45:45 +08:00
parent 0ff8b2353f
commit c8037746c3
2 changed files with 30 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
README.md
View File

@ -2,7 +2,7 @@
* @Author: LiangSong(sl12160010@gmail.com)
* @Date: 2023-03-10 21:18:35
* @LastEditors: LiangSong(sl12160010@gmail.com)
* @LastEditTime: 2023-04-28 19:52:27
* @LastEditTime: 2023-04-28 22:44:21
* @FilePath: /Open-Llama/README.md
* @Description:
*
@ -22,9 +22,21 @@ Open-Llama是一个开源项目，提供了一整套用于构建大型语言模
**训练速度达到3620 token/s，快于Llama原文中的3370 token/s，达到目前sota的水平。**
经过Instruct-tuning的CheckPoint已开源在[s-JoL/Open-Llama-V1](https://huggingface.co/s-JoL/Open-Llama-V1)。使使用ckpt需要先用下面命令安装最新版本Transformers
``` base
经过Instruct-tuning的CheckPoint已开源在[HuggingFace: s-JoL/Open-Llama-V1](https://huggingface.co/s-JoL/Open-Llama-V1)。使用ckpt需要先用下面命令安装最新版本Transformers
``` python
pip install git+https://github.com/s-JoL/transformers.git@dev
from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("s-JoL/Open-Llama-V1", use_fast=False)
model = AutoModelForCausalLM.from_pretrained("s-JoL/Open-Llama-V1").cuda()
inputs = tokenizer('user:implement quick sort in python\nsystem:', return_tensors='pt', return_attention_mask=False)
for k, v in inputs.items():
inputs[k] = v.cuda()
pred = model.generate(**inputs, max_new_tokens=512, do_sample=True)
print(tokenizer.decode(pred.cpu()[0]).strip())
```
只经过预训练的CheckPoint也上传至[s-JoL/Open-Llama-V1-pretrain](https://huggingface.co/s-JoL/Open-Llama-V1-pretrain)。
模型已提交[PR](https://github.com/huggingface/transformers/pull/22795)合并至Transformers main分支。

18
README_en.md
View File

@ -2,7 +2,7 @@
* @Author: LiangSong(sl12160010@gmail.com)
* @Date: 2023-03-10 21:18:35
* @LastEditors: LiangSong(sl12160010@gmail.com)
* @LastEditTime: 2023-04-28 19:53:01
* @LastEditTime: 2023-04-28 22:44:27
* @FilePath: /Open-Llama/README_en.md
* @Description:
*
@ -22,9 +22,21 @@ Open-Llama is an open-source project that offers a complete training pipeline fo
**The training speed reaches 3620 tokens/s, faster than the 3370 tokens/s reported in the original Llama paper, reaching the current state-of-the-art level.**
The CheckPoint after Instruct-tuning is open-source on [s-JoL/Open-Llama-V1](https://huggingface.co/s-JoL/Open-Llama-V1). To use the CheckPoint, first, install the latest version of Transformers with the following command:
``` base
The CheckPoint after Instruct-tuning is open-source on [HuggingFace: s-JoL/Open-Llama-V1](https://huggingface.co/s-JoL/Open-Llama-V1). To use the CheckPoint, first, install the latest version of Transformers with the following command:
``` python
pip install git+https://github.com/s-JoL/transformers.git@dev
from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("s-JoL/Open-Llama-V1", use_fast=False)
model = AutoModelForCausalLM.from_pretrained("s-JoL/Open-Llama-V1").cuda()
inputs = tokenizer('user:implement quick sort in python\nsystem:', return_tensors='pt', return_attention_mask=False)
for k, v in inputs.items():
inputs[k] = v.cuda()
pred = model.generate(**inputs, max_new_tokens=512, do_sample=True)
print(tokenizer.decode(pred.cpu()[0]).strip())
```
The CheckPoint after pre-training only is also uploaded to [s-JoL/Open-Llama-V1-pretrain](https://huggingface.co/s-JoL/Open-Llama-V1-pretrain).
The model [PR](https://github.com/huggingface/transformers/pull/22795) has been submitted for merging into the Transformers main branch.