Installation#

Bio-transformers can be installed in Python 3.7 and external python dependencies are mainly defined in requirements. There are multiple different methods to install Bio-transformers:

  1. Clone Bio-transformers and create a virtual environment using Anaconda / Miniconda (recommended).

  2. Clone Bio-transformers and build a docker image using the provided docker file. (not implemented)

  3. Install directly from PyPI release without cloning Bio-transformers.

Install torch/cuda#

Warning

bio-transformers doesn’t manage the installation of cuda toolkit and torch gpu version.

If you want to find a specific version or torch based on your CUDA setup, please refer to this page

The Dockerfile provided in the github repository relies on :

  • pip install torch==1.8.0+cu111 torchvision==0.9.0+cu111 torchaudio==0.8.0 -f https://download.pytorch.org/whl/torch_stable.html

Install in conda environment#

The recommended method is to install Bio-transformers in a dedicated virtual environment using Anaconda / Miniconda.

conda create --name bio-transformers python=3.7 -y
conda activate bio-transformers
pip install bio-transformers