I have started to use the H2O library and I am now going through an R script that I am "translating" over to python. I want to perform these partial dependency plots, however, I don't quite understand what is the output of these plots.
For context, my ML model is a classification model and has 2 classes. Now, when I perform this plot, my mean response is the mean of the output (0 and 1) or the mean of the probability of being 1 (p1)?
My output graph looks like this:
And my code looks like this:
dados_h2o_duracao = h2o.H2OFrame(df_copy[(df_copy['duracao_apolice']>=18) & (df_copy['duracao_apolice']<=80)])
duracao = model.partial_plot(data=dados_h2o_duracao,cols=['duracao_apolice'])