README.md 1.99 KB
Newer Older
shenggan's avatar
shenggan committed
1
# FastFold
Shenggan's avatar
Shenggan committed
2

zhuwenwen's avatar
zhuwenwen committed
3
4
5
## 安装
安装FastFold,你需要
+ Python 3.8 或者 3.9.
Shenggan's avatar
Shenggan committed
6

zhuwenwen's avatar
zhuwenwen committed
7
### 使用pip方式安装
zhuwenwen's avatar
zhuwenwen committed
8
9
fastfold whl包下载目录:[https://cancon.hpccube.com:65024/4/main/fastfold/dtk23.04](https://cancon.hpccube.com:65024/4/main/fastfold/dtk23.04)
根据对应的pytorch版本和python版本,下载对应fastfold的whl包
Shenggan's avatar
Shenggan committed
10

Fazzie-Maqianli's avatar
Fazzie-Maqianli committed
11
```shell
zhuwenwen's avatar
zhuwenwen committed
12
pip install fastfold* (下载的fastfold的whl包)
Shenggan's avatar
Shenggan committed
13
14
```

zhuwenwen's avatar
zhuwenwen committed
15
### 使用源码编译方式安装
16

zhuwenwen's avatar
zhuwenwen committed
17
#### 编译环境准备
zhuwenwen's avatar
zhuwenwen committed
18
19
20
21
22
提供2种环境准备方式:

1. 基于光源pytorch基础镜像环境:镜像下载地址:[https://sourcefind.cn/#/image/dcu/pytorch](https://sourcefind.cn/#/image/dcu/pytorch),根据pytorch、python、dtk及系统下载对应的镜像版本。

2. 基于现有python环境:安装pytorch,pytorch whl包下载目录:[https://cancon.hpccube.com:65024/4/main/pytorch/dtk23.04](https://cancon.hpccube.com:65024/4/main/pytorch/dtk23.04),根据python、dtk版本,下载对应pytorch的whl包。安装命令如下:
Shenggan's avatar
Shenggan committed
23
```shell
zhuwenwen's avatar
zhuwenwen committed
24
pip install torch* (下载的torch的whl包)
zhuwenwen's avatar
zhuwenwen committed
25
pip install setuptools==59.5.0 wheel
26
27
```

zhuwenwen's avatar
zhuwenwen committed
28
29
#### 源码编译安装
- 代码下载
30
```shell
zhuwenwen's avatar
zhuwenwen committed
31
git clone https://developer.hpccube.com/codes/aicomponent/fastfold  # 根据需要的分支进行切换
32
33
```

zhuwenwen's avatar
zhuwenwen committed
34
35
36
- 提供2种源码编译方式(进入fastfold目录):
```
1. 编译whl包并安装
zhuwenwen's avatar
zhuwenwen committed
37
export CPLUS_INCLUDE_PATH=$ROCM_PATH/include:$ROCM_PATH/cuda/targets/x86_64-linux/include:$CPLUS_INCLUDE_PATH
zhuwenwen's avatar
zhuwenwen committed
38
python setup.py bdist_wheel 
zhuwenwen's avatar
zhuwenwen committed
39
pip install dist/fastfold*
zhuwenwen's avatar
zhuwenwen committed
40

zhuwenwen's avatar
zhuwenwen committed
41
42
43
2. 源码编译安装
export CPLUS_INCLUDE_PATH=$ROCM_PATH/include:$ROCM_PATH/cuda/targets/x86_64-linux/include:$CPLUS_INCLUDE_PATH
python3 setup.py install 
Shenggan's avatar
Shenggan committed
44
45
```

zhuwenwen's avatar
zhuwenwen committed
46
## 版本号查询
zhuwenwen's avatar
zhuwenwen committed
47
- python -c "import fastfold; print(fastfold.\_\_version__)",版本号与官方版本同步,查询该软件的版本号,例如0.2.0;
zhuwenwen's avatar
zhuwenwen committed
48

zhuwenwen's avatar
zhuwenwen committed
49
50
## Known Issue
-
zhuwenwen's avatar
zhuwenwen committed
51

zhuwenwen's avatar
zhuwenwen committed
52
53
54
## Note
+ 若使用 pip install 下载安装过慢,可添加源:-i https://pypi.tuna.tsinghua.edu.cn/simple/
+ ROCM_PATH为dtk的路径,默认为/opt/dtkxxx
Shenggan's avatar
Shenggan committed
55

zhuwenwen's avatar
zhuwenwen committed
56
57
## 参考
- [README_ORIGIN](README_ORIGIN.md)