As Machine Learning tools become mainstream, and ever-growing choice of these is available to data scientists and analysts, the need to assess those best suited becomes challenging. In this study, 20 Machine Learning models were benchmarked for their accuracy and speed performance on a multi-core hardware, when applied to 2 multinomial datasets differing broadly in size and complexity.
Suggestions for additional Machine Learning, pertinent datasets and which recommender to benchmark are welcome. The full analysis can be found here.
None of these examples seem to have the data separated into training and test sets. Could you please share where this step is taken in the code in the linked full analysis? Otherwise, are we to assume these reported accuracy numbers represent only observations the models were trained on?