Other Packages¶

There are lots of other packages in the python ecosystem.

This is an opinionated list with packages sorted in order of preference, and/or the best order for learning.

Note

If the research group has a particular interest in any of the following packages. Please add a comment to this issue on the workshop repository Issues list and let me know as a comment

Plotting and Visualisation¶

Package	Description
matplotlib	The core library for generating plots
seaborn	A plotting library with a focus on statistical graphics
bokeh	build plots for the web through javascript powered visualisations
Plotly	a `python` interface to Plotly for interactive plots

Statistical Modelling & Machine Learning¶

Package	Description
statsmodels	Statistical modelling (Econometrics focus)
linearmodels	Extends `statsmodels` to include Panel Models, High Dimensional Fixed Effects, System Regressions etc.
scikit-learn	The standard machine learning library
Stan	Bayesian Statistical modeling, data analysis, and prediction
PyMC3	Alternative package for Bayesian statistical modeling
scikit-survival	Survival analysis based on top of scikit-learn
lifelines	Survival analysis in Python

Performance & Scale¶

Package	Description
Numba	Generates optimised code using a `just-in-time` compiler
Dask	Infrastructure for building parallel workloads across `cpu cores` or `machines`
Cython	Translates Python-style code to C

Scientific Computing & Mathematics¶

Package	Description
scipy	The core library for `scientific computing` including statistics, linear algebra, numerical integration and optimisation
cvxpy	Convex optimisation
FEniCS	Platform for partial differential equations (PDE)

Networks¶

Package	Description
networkx	NetworkX is a Python package for the creation, manipulation, and study of the structure, dynamics, and functions of complex networks
networkit	A toolkit for large-scale network analysis
graph-tool	Network library designed to be fast and largely implemented in C/C++
Snap.py	A python interface for SNAP (a general purpose, high performance library for analysis and manipulation of large networks)
DeepGraph	A `pandas` based entry point into defining networks

The article Python Packages for Networks is a good reference [1].

Others:

igraph,
tulip

Natural Language Processing (NLP)¶

Course	Description
NLTK	Natural Language Toolkit
SpaCy	Library with lots of NLP algorithms
gensim	Topic modelling

Online Courses¶

Course	Description
Python Programming for Economics and Finance	Introduction to the `python` language
Quantitative Economics with Python	A series of lectures on quantitative economic modelling
Software Carpentry	Introduction to various programming topics
Data carpentry	Hands-on, interactive lessons on various data analysis topics

1: Vladimir Batagelj. Python Packages for Networks, pages 1943–1952. Springer New York, New York, NY, 2018. URL: https://doi.org/10.1007/978-1-4939-7131-2_110210, doi:10.1007/978-1-4939-7131-2_110210.

RSIT Workshop (Uni Tübingen) 2021