Conda environment tutorial

$README.md

Create conda environment

1. Download and install miniconda3: https://docs.conda.io/en/latest/miniconda.html
2. Open Anaconda Prompt (miniconda3)
3. conda create --prefix=conda-env python==3.8
4. conda config --set env_prompt '({name}) <-- Just once
5. conda activate conda-env
6. conda install <package-name>
7. conda install -c conda-forge <package-name>
8. Inside conda-env, run: pip install <package-name>

Create an insolated conda environment

Conda envs aren’t globally isolated. The default behavior when creating a pip virtual environment is that when inside the new environment, you do not have access to globally installed pip packages. If you want access to global packages, you need to initialize your virtual environment with:

virtualenv env --system-site-packages

conda , on the other hand, lets you access global system packages by default. If you want the environment to be completely isolated from these global packages like with pip, you can create the conda environment based on a clone of an empty pip virtualenv.

1. pip install virtualenv
2. conda create --prefix=tmp-env python==3.8
3. conda activate /tmp-env
4. virtualenv empty-env <-- run this command with the tmp-env activated
5. conda create --prefix=conda-env --clone empty-env <-- this will create an insolated conda-env with python==3.8

Note

Steps 4 and 5 in the insolated environment are necessary since virtualenv uses the default python interpreter in the current path.

References

• https://towardsdatascience.com/a-guide-to-conda-environments-bc6180fc533
• https://towardsdatascience.com/guide-to-conda-for-tensorflow-and-pytorch-db69585e32b8
• https://gist.github.com/Geoyi/d9fab4f609e9f75941946be45000632b

Run conda init beforehand

make sure run conda init powershell in windows