Interpreting the scores

In this section, we will display the scores achieved by the five models. You can try to improve the performance of a model by fine-tuning the parameters.

You can also add several other models to measure their performance.

In this chapter, we will focus on the LIME model-agnostic explainer and let the program select the model that produces the best performance.

The program displays the AutoML experiment summary:

# @title AutoML experiment: Summary
print("The best model is", clf, "with a score of:", round(best, 5))
print("Scores:")
print("Random forest         :", round(score1, 5))
print("Bagging               :", round(score2, 5))
print("Gradient boosting     :", round(score3, 5))
print("Decision tree         :", round(score4, 5))
print("Extra trees           :", round(score5, 5))

The output displays the summary:

The best model is Extra Trees with a...