fpp-analysis-tools

Collection of tools designed to analyse time series of intermittent fluctuations.

Installation

The package is published to PyPI and can be installed with

pip install git+https://github.com/uit-cosmo/fpp-analysis-tools

If you want to contribute to the project you must first clone the repo to your local machine, then install the project using poetry:

git clone [email protected]:uit-cosmo/fpp-analysis-tools.git
cd fpp-analysis-tools
poetry install

If you plan to use the GPUs, specifically useful for the deconvolution, (local) installation using either pixi or conda is supported (the conda environment file is exported by pixi as pixi project export conda-environment environment.yml):

git clone [email protected]:uit-cosmo/fpp-analysis-tools.git
cd fpp-analysis-tools
# pixi
pixi install
# conda
conda env create --name name-of-my-env --file environment.yml

Troubleshooting

There is a chance you face a Failed to import CuPy. issue, or that the libfile is not found (see a more thorough walk-through here).

Check if you can find the file libcublas.so.10 (or libcuda.so or similar) in /usr/lib/ or any of its subdirectories. (For example find /usr/lib/ -name 'libcublas.so*', or with fd fd libcublas.so /usr/lib). On the machine used by our group, the following extra step is necessary after installing the project:

export LD_LIBRARY_PATH="/usr/lib/x86_64-linux-gnu/"

Usage

You can import all functions directly from fppanalysis, such as

import fppanalysis as fa

bin_centers, hist = fa.get_hist(Data, N)