This is a repository with some useful snippets, tutorial or explaination linked to computation in python.
Broad topics will be talked about (and the repo is under construction).
Deps can be installed with Anaconda
with conda env create in the root folder.
You can then activate the environment and launch Jupyter with
conda activate locie-notebook
jupyter notebookUseful tutorial or snippets about the python language itself : data structure, useful libraries not directly related to numerical computation.
- Unlock the power of your computer with multiprocessing computation (en)
- !DRAFT functools and functional programming (en)
- !DRAFT functools et la programmation fonctionnelle (fr)
- !DRAFT Classy iterable with itertools (en)
- !DRAFT Itérer avec classe avec itertools (fr)
Numpy is the main brick of the computation in python. It's based on C routines which allow fast computation without have to sacrifice the ease of use.
Scipy is a higher level brick than numpy and contain a lot of powerful routines to deal with almost all numerical analysis cases : interpolation, signal analysis, integration, linear algebra, statistic analysis and so on.
Pandas is the dedicated tool to deal with tabular data. It's the excel on steroid for python. It allows easy manipulation and complex request on the data, statistic analysis, a broad range of I/O tool that allows to import a lot of different file formats. (.csv, .xls/x, .hdf, SQL databases, for multiple files, remote data...).
Regroup all the viz tools: the classic matplotlib, but also seaborn (statistic based vizualisation), bokeh and holoviews that allow rich and interactive vizualisation.
Is a set of quick tutorial that introduce to machine learning via the sklearn library or to the optimization tool we use at LOCIE.
Other topic that does not fit into other folders.