Python¶

Powerful programming language, well-maintained by development team.

Versatile! Can be used as a simple calculator, to organize and manipulate data structures, visualize data, for machine learning, and even manipulate MRI data.
Macs come pre-installed with python 2.7 but most packages will likely require python 3(+).
To take advantage of the robustness of python, recommend installing libraries.
- Library = package that uses python to perform pre-set functions
Some libraries are built upon others–they add functionality to already existing functions.
Python documentation.
Resources for learning how to use Python for data science.

Installation¶

To install the latest version of Python, go to Python for Mac, download the macOS 64-bit installer, and follow prompts.

Data Organization & Processing

Pandas: organize, manipulate, plot single column series data, and data frames tables
NumPy: short for Numerical Python, perform operations on arrays and matrices
SciPy: conduct linear algebra, integration, optimization, and statistics

Data Visualization

Matplotlib: plotting, making pretty figures (line graphs, scatterplots, histograms)
Seaborn: extends Matplotlib for making figures to visualize statistical tests and models (correlation matrices, connectivity matrices, distributions)
Plotly: web-based tool; graph plotting library for creating and displaying figures, hovering over to show details.
Bokeh: for interactive and scalable visualizations through web-browsers (aka cool presentations).

Machine Learning/AI

Scikit-Learn: group of packages in the SciPy stack
- Machine learning and data mining. Think classification, clustering, regression, dimensionality reduction, model selection.
TensorFlow: artificial intelligence library for creating large-scale neural networks (machine learning + deep learning)
Keras: neural network library
- TensorFlow’s high-level API for building and training Deep Neural Network code
- Statistical modeling but with images + text