daxs 2022.0

Data Analysis for X-ray Spectroscopy

Data Analysis for X-ray Spectroscopy

Installation

You can install the development version using: python3 -m pip install [--ignore-installed] https://gitlab.esrf.fr/spectroscopy/daxs/-/archive/main/daxs-main.tar

The --ignore-installed argument is required if you want to upgrade an existing installation.

It is best if you install the library in a virtual environment to avoid messing up other Python packages. See the official documentation on how to create and use virtual environments.

Usage at the ESRF

Scripts and command line

To use the library on the computing cluster, follow the steps below.

1. Login on the front-end: ssh -Y slurm-nice-devel
2. Ask for resources: srun --pty bash
3. Load the spectroscopy module: module load conda; module load spectroscopy. The command loads a Python virtual environment that contains the latest version of the library.
4. Print the version of the library to test that everything went smoothly: python -c "import daxs; print(daxs.__version__)"

If all went well, you should be able to use the library in your scripts.

JupyterHub

You can also use the library in a Jupyter Notebook. Follow steps 1-3 from above. Next, install the kernel by running kernel-install.

After, connect to https://jupyter-slurm.esrf.fr and start a server on the Intel partition. On the right-hand side, press New. The Python environment should be in the list, as shown in the image below. Select it to create a new Jupyter Notebook.

Run kernel-remove to remove the kernel. Alternatively, you can use the jupyter kernelspec command to manage the kernels.

While this simplifies the usage, you will not be able to add Python packages to the virtual environment. If you want to use additional packages not present in the environment, the best approach is to install the library in your home directory, in a virtual environment.

Documentation

The documentation can be found at https://spectroscopy.gitlab-pages.esrf.fr/daxs.