Python Data Analysis Visualisations

I have been enormously impressed with Python in general, but one of its many strengths is just how quick and easy some of its libraries make generating just the right visualisations to aid in exploratory data analysis.

Data analysis work was carried out on the melodies of folk songs and Beatles songs for an algorithmic music project. Digraphs and plots were generated to show growth over time in terms of pitch-duration symbol nodes and the arc transitions between them. Verse melody notes are in red, chorus melody notes are in blue and rests are in green. The first derivatives of the line plots are shown as small bar charts.

Another project where Matplotlib was particularly helpful was a backpropagation classification neural network for clustering test data based on clusters learnt from training data. Cluster centroids are shown separately in the plots on the right hand side of the figure.

But don't worry too much about the details and the applications here, just enjoy the visual representations!


Visualisations used in data analysis, and generated with Python and Graphviz/Matplotlib:




The selected visualisation: