Segmented loss

by: Valdemar Stentoft-Hansen & Chris Cave
(Feyn version 2.0 or newer)

A Models performance varies across a dataset. The plot_segmented_loss displays the distribution of an input or output as a histogram and is overlayed with the average loss for the associated bin.

You can use this to determine areas where the Model performs best and worst across the inputs and output.

import feyn

from sklearn.datasets import load_boston
import pandas as pd

from feyn.tools import split

#Download boston housing dataset
boston = load_boston()
df_boston = pd.DataFrame(boston.data, columns=boston.feature_names)
df_boston['PRICE'] = boston.target

# Train/test split
train, test = split(df_boston)

# Connect to QLattice
ql = feyn.connect_qlattice()

models = ql.auto_run(
    data=train,
    output_name='PRICE'
)
# Select the best Model
best = models[0]

best.plot_segmented_loss(
    data = train,
    by = None
)

data

The data passed in this parameter is used to determine the bins of the input or output the plot is segemented by and the average loss across the bin.

by

The input name or output to segment the loss over.

Saving the plot

You can save the plot using the filename parameter. The plot is saved in the current working directory unless another path specifed.

best.plot_segmented_loss(train, filename="feyn-plot")

Location in Feyn

This function can also be found in feyn.plots module.

from feyn.plots import plot_segmented_loss

plot_segmented_loss(best, train)