Installation

Before you begin, it’s highly recommended to set up a Python virtual environment for mesalab. Virtual environments create an isolated space for your project’s dependencies, preventing conflicts with other Python projects or your system’s global Python installation. This ensures a clean and reproducible setup. You can choose your favorite virtual environment manger.

If you’re not familiar with this process, here are two commonly used and recommended options: venv (Python’s built-in virtual environment module) or Conda (a popular package, dependency, and environment manager).

Using venv (Recommended for most Python-only projects)

To create and activate a virtual environment named mesalab-env using venv, navigate to your project directory and run:

python3 -m venv mesalab-env
source mesalab-env/bin/activate

You’ll know the virtual environment is active when you see (mesalab-env) preceding your prompt in the terminal.

Using Mamba or Conda (Recommended for complex environments)

Standard Conda can be slow when resolving large scientific stacks. It is highly recommend using Mamba (a fast, C++ based drop-in replacement for Conda) to drastically speed up the installation.

If you have Mamba or Conda installed, you can create a dedicated environment for mesalab:

Create the environment:

Option A: Using Mamba (Fastest)

mamba create -n mesalab-env python=3.9
mamba install -c conda-forge numpy pandas matplotlib scipy pyyaml tqdm numba swifter dask pyarrow h5py astropy addict

Option B: Using Conda (Legacy)

conda create -n mesalab-env python=3.9
conda install -c conda-forge numpy pandas matplotlib scipy pyyaml tqdm numba swifter dask pyarrow h5py astropy addict

Activate the environment:
```
conda activate mesalab-env
```
You’ll know the environment is active when you see (mesalab-env) preceding your prompt in the terminal.

Once your virtual environment is active, you can proceed with installing the necessary dependencies.

Install from pip

To install mesalab with pip:

$ pip install mesalab

Building from source

You can also install the current development version of mesalab with cloning the GitHub repository and install the code with pip:

$ git clone https://github.com/konkolyseismolab/mesalab
$ cd mesalab
$ pip install .

Unit test

To run the unit tests you may need to install pytest:

$ pip install pytest
$ python3 -m pytest