installation.rst

Installation guide

Table of Contents

Prerequisites

• OS support: Windows, Linux and OSX
• Python version: 3.6, 3.7

Dependencies

(- means no specific version requirement for that package)

Name	Version
numpy	-
pandas	-
networkx	-
paddlepaddle	>=2.0.0rc0
pgl	>=2.1
rdkit	-
sklearn	-

Instruction

Since PaddleHelix depends on the paddlepaddle of version 2.0.0rc0 or above, and rdkit cannot be installed directly using pip, we suggest using conda to create a new environment for the installation. Detailed instruction is shown below:

• If you do not have conda installed, please install it at first:
• Create a new environment with conda:

$ conda create -n paddlehelix python=3.7

• Activate the environment just created:

$ conda activate paddlehelix

• Install rdkit using conda:

$ conda install -c conda-forge rdkit

• Install the right version of paddlepaddle according to the device (CPU/GPU) you want to run PaddleHelix on.

1. If you want to use the GPU version of paddlepaddle, run this:

$ python -m pip install paddlepaddle-gpu -f https://paddlepaddle.org.cn/whl/stable.html

2. Or if you want to use the CPU version of paddlepaddle, run this:

$ python -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple

Note

The version of paddlepaddle should be higher than 2.0. Check paddlepaddle official document for more installation guide.

• Install pgl using pip:

$ pip install pgl

• Install PaddleHelix using pip:

$ pip install paddlehelix

• The installation is done!

Note

After playing, if you want to deactivate the conda environment, do this:

$ conda deactivate