Installation

Prerequisites

Conda Environment

We recommend using Anaconda / Miniconda to create a conda environment for using CPA. You can create a python environment using the following command:

conda create -n cpa python=3.9

Then, you can activate the environment using:

conda activate cpa

Pytorch

CPA is implemented in Pytorch and requires Pytorch version >= 1.13.1.

OSX

You can install Pytorch 1.13.1 using the following command:

pip install torch==1.13.1

Linux and Windows

If you have access to GPUs, you can install the GPU version of Pytorch following the instructions here .

Sample command for installing Pytorch 1.13.1 on different CUDA versions:

# ROCM 5.2 (Linux only)

pip install torch==1.13.1+rocm5.2 –extra-index-url https://download.pytorch.org/whl/rocm5.2

# CUDA 11.6

pip install torch==1.13.1+cu116 –extra-index-url https://download.pytorch.org/whl/cu116

# CUDA 11.7

pip install torch==1.13.1+cu117 –extra-index-url https://download.pytorch.org/whl/cu117

# CPU only

pip install torch==1.13.1+cpu –extra-index-url https://download.pytorch.org/whl/cpu

Installing CPA

Finally, You can install latest version of CPA through our Github:

pip install git+https://github.com/theislab/cpa

Dependencies

Install scanpy afterwards:

pip install scanpy

Colab

If working on Google Colab, run the following cell at the beginning:

import sys
#if branch is stable, will install via pypi, else will install from source
branch = "latest"
IN_COLAB = "google.colab" in sys.modules

if IN_COLAB and branch == "stable":
    !pip install cpa-tools
    !pip install scanpy
elif IN_COLAB and branch != "stable":
    !pip install --quiet --upgrade jsonschema
    !pip install git+https://github.com/theislab/cpa
    !pip install scanpy