I am excited to share a new deep learning model performance trajectory graph.

Here is an example produced based on Keras in R using ggplot2:

The ideas include:

• We plot model performance as a function of training epoch, data set (training and validation), and metric.
• For legibility we facet on metric, and facets are adjusted so all facets have the same visual interpretation (“up is better”).
• The only solid horizontal curve is validation performance, and training performance is only indicated as the top-region of a shared region that depicts degree of over-fit.

Obviously is going to take some training and practice to read these graphs quickly: but that is petty much true for all visualizations.

The methods work with just about any staged machine learning algorithm (neural nets, deep learning, boosting, random forests, and more) and can also be adapted to non-staged bug regularized methods (lasso, elastic net, and so on).

The graph is now part of the development version of WVPlots. And we have complete worked examples for Keras and xgboost.