Visualization of EnsemblePursuit

I mentioned in a previous post that I am participating in the AlgoExpert Software Engineering Project competition.

This is today's work. I used pyqtgraph's OpenGL framework to visualize the cost function of adding neurons to an ensemble using the Ensemble Pursuit algorithm that my boss invented. If you want to learn more about the algorithm, here's the github page https://github.com/MouseLand/EnsemblePursuit Essentially, the algorithm groups together neurons that explain each other's variance. It's a greedy algorithm that iteratively adds neurons to an ensemble, while the variance explained of the neuron by the ensemble is within a threshold (when it hits a threshold it stops and moves on to fitting the next ensemble). 

Now I need to make scroll bar that re-renders the plot above for each iteration, effectively showing the cost function of the algorithm. 

It took me 8 hours to make this plot:-) pyqtgraph and OpenGL are awesome. Check out this project on github https://github.com/mariakesa/EnsemblePursuitViz/blob/master/Notebooks/FirstAttempt.ipynb