# alphafold安裝(anaconda) ## 安裝anaconda [anacondarul教學](https://zhuanlan.zhihu.com/p/64930395) 首先下載anaconda ``` wget https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/Anaconda3-2019.03-Linux-x86_64.sh ``` or ``` yum -y install wget ``` 安裝anaconda ``` bash Anaconda3-2019.03-Linux-x86_64.sh ``` 一直按yes、enter 出現Thank you for installing Anaconda3!即安裝好 啟用 ``` source activate ``` 如果失敗 ``` vi /etc/profile ``` 在文件最后加入如下语句（路径需要根据自己的安装位置更改） ``` PATH=$PATH:/opt/anaconda3/bin export PATH ``` 按住shift键+:键，输入wq，保存文件并退出 ``` source /etc/profile ``` ``` echo $PATH ``` 在啟用一次應該可以 ``` source activate # 进入conda环境 出现(base)则说明安装成功 source deactivate # 退出conda环境 ``` ## 安裝Dependencies * Python 3.6+(安裝完anaconda就會有) * Abseil 0.8.0 ``` pip3 install absl-py conda install -c anaconda absl-py==0.8.0 ``` 測試Abseil ``` git clone https://github.com/abseil/abseil-py.git cd abseil-py ``` * Numpy 1.16(conda版)中間要按Y ``` conda create -n my-env conda activate my-env conda config --env --add channels conda-forg conda install numpy ``` * Six 1.12 ``` pip3 install six ``` * Setuptools 41.0.0 ``` pip3 install --upgrade setuptools ``` * Sonnet 1.35(anaconda) ``` conda install -c conda-forge dm-sonnet==1.35 conda install -c anaconda tensorflow-gpu ``` * TensorFlow 1.14(anaconda) ``` conda install -c conda-forge tensorflow=1.14 ``` * TensorFlow Probability 0.7.0 ``` pip install tensorflow-probability==0.7.0 ``` 進入python ``` import tensorflow as tf import sonnet as snt print("TensorFlow version {}".format(tf.__version__)) print("Sonnet version {}".format(snt.__version__)) ``` 完成