Skip to main content

Quick-install/provision a fresh Miniconda distribution from scratch.

Project description

A script to quickly make/provision a fresh Miniconda installation from scratch.

The goal of this script is to quickly install a fresh Miniconda across different operating systems and use it in a CI/CD context, too. It will download a Miniconda binary based on your operating system, unpack the binary and install it locally. It was tested on MacOS and Linux ok, has decent Windows support and also aims to support ARM6 and ARM7 on the Raspberry Pi via Berryconda (sadly no longer maintained).

This script has no external dependencies, but expects to be run with Python 3.6+.

N.B. This project is often updated online which is why its git history might look strange…


You can use pip (for Python 3) to install and uninstall this package like the following:

pip install mcinstall
pip uninstall mcinstall

More recent, not yet released, versions can be installed like the following without cloning the repo:

pip install -e git+

Or you clone the repo locally and install the package like this:

git clone
cd mcinstall
pip install -e .

Sample Usage

If you run it like this:

mcinstall ~/Downloads/mc3

it will run a command like this on macOS (using some defaults and decent system introspection) to create a conda base installation:

bash -b -f -p ~/Downloads/mc3

This can be activated then with a command like this:

source ~/Downloads/mc3/bin/activate

Windows Example:

On Windows if you execute this command:

mcinstall %USERPROFILE%\downloads\mc3

it will run a command like this:

start /wait "" Miniconda3-latest-Windows-x86_64.exe /InstallationType=JustMe /RegisterPython=0 /S /D=%USERPROFILE%\downloads\mc3

This conda installation can be activated then with a command like this:


Suggested test:

mcinstall --verbose --pip-dependencies jupyter,torch ~/Downloads/torchy
source ~/Downloads/torchy/bin/activate
python -c "import torch; print('ok')"

Suggested test dependencies files:

$ more ~/Downloads/reqs.txt

$ more ~/Downloads/env.yml
name: test
  - conda-forge
  - voila

Sample run (replaced home directory with ~ manually):

$ mcinstall --verbose --pip-dependencies jupyter,torch ~/Downloads/torchy
Making directory ~/Downloads/torchy.
Downloading ...
Copying to ~/Downloads/ ...
Running command: bash ~/Downloads/ -b -f -p ~/Downloads/torchy
Unpacking payload ...
Collecting package metadata (current_repodata.json): ...working... done
Solving environment: ...working... done


Preparing transaction: ...working... done
Executing transaction: ...working... done
installation finished.

Running command: ~/Downloads/torchy/bin/pip install jupyter

Running command: ~/Downloads/torchy/bin/pip install torch

Run this to start using your fresh Miniconda: "source ~/Downloads/torchy/bin/activate".

Sample log file (mcinstall.log):

mv ~/Downloads/
bash ~/Downloads/ -b -f -p ~/Downloads/torchy
source ~/Downloads/torchy/bin/activate
~/Downloads/torchy/bin/pip install jupyter
~/Downloads/torchy/bin/pip install torch


  • improve logging executed commands to reproduce them as a shell script
  • add self.clean_dest_path to MinicondaInstaller.__init__
  • same with self.download_path
  • consolidate Windows support
  • improve configuration
  • make a sample screencast with asciinema

Project details

Release history Release notifications

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for mcinstall, version 0.3.0
Filename, size File type Python version Upload date Hashes
Filename, size mcinstall-0.3.0.tar.gz (7.3 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page