The accuracy of the classification model for a random sample is evaluated according to the results when the model is and is not used. The Gains/Lift Table page uses predicted data to evaluate model performance. More details on gains and lift table in H2O can be found here.
If you are looking for how to get these model metrics from H2O model in python you can look at here.
When looking at the gains and lift table for a classification model in H2O Flow you will see the results as below:
The gains and lift table shows results into 16 buckets and the buckets are fixed for the following values.
[0.01, 0.02, 0.03, 0.04, 0.05, 0.10, 0.15, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1]
Here are some of the python code snippets you can use to try it out:
mymodel = h2o.get_model("gbm_pojo_test") mymodel.gains_lift() mymodel.gains_lift()['group']
Thats all, enjoy!!