Use Anaconda to Organize Python Environment

Working on multiple Python projects can quickly become chaotic if you don't have a good system for managing dependencies and isolating environments.

2 min readMar 17, 2024

That's where Anaconda's environment management capabilities shine. Anaconda is a popular Python distribution that includes conda, a powerful package and environment manager.

With conda, you can create separate, isolated Python environments for each project. Each environment can have its own set of installed packages and dependencies without interfering with other environments or your system's Python installation. This isolation is a game-changer for maintaining clean, reproducible, and conflict-free development setups.

Creating a New Environment

Setting up a new conda environment is straightforward. Simply use the `conda create` command, specify a name for your environment, and optionally set the Python version:

conda create --name myproject python=3.9

This creates a Python 3.9 environment named "myproject". To start using this environment, you activate it:

conda activate myproject

Your terminal prompt will reflect the activated environment. Any Python packages you install will be isolated to this "myproject" environment.

Listing Environments

As you create more environments, you can list them all using:

conda env list

This will show all your conda environments and indicate which one is activated.

Installing Packages

With an environment activated, you can install packages using either conda installor pip install. Conda will first check its package channels to see if a conda package is available. If not, it will defer to PyPI:

conda install numpy           # Installs NumPy from conda channel 
pip install requests          # Installs requests from PyPI

Install From Requirement.txt

In case you need to install the package through the requirement.txt file.

#For requirement.txt
#conda
conda install --file requirements.txt
#pip
#If some packages are not found in the Anaconda Cloud channels, 
#you can install them using pip within the activated conda environment:
pip install -r requirements.txt

Reproducing Environments

One of the biggest benefits of conda environments is reproducibility. You can use a requirement.txt file to capture all dependencies:

conda create --name myenv --file requirements.txt

This installs all packages in the requirement.txtfile into a newly created "myenv" environment. Alternatively, you can send a requirement.txtto a colleague and they can exactly reproduce your environment.

Switching Between Environments

With multiple environments, you can easily switch between them as needed using conda activate:


conda activate myproject      # Activate myproject environment
conda activate datasci        # Switch to datasci environment

When you're done, deactivate to return to your base environment.

conda deactivate

Conclusion

Conda makes it easy to keep your Python setups clean, isolated, and reproducible. Give it a try on your next Python project to streamline your development workflow.

Follow for more