Feyn Documentation

Feyn Documentation

  • Learn
  • Guides
  • Tutorials
  • API Reference
  • FAQ

›Evaluate Regressors

Getting Started

  • Quick start
  • Using Feyn
  • Installation
  • What is the QLattice?

Essentials

  • Auto Run
  • Summary plot
  • Plot response
  • Splitting a dataset
  • Seeding a QLattice
  • Predicting with a model
  • Saving and loading models
  • Categorical features

Evaluate Regressors

  • Regression plot
  • Residuals plot

Evaluate Classifiers

  • ROC curve
  • Confusion matrix
  • Plot probability scores

Understand Your Models

  • Plot response 1D
  • Plot response 2D
  • Model signal
  • Segmented loss
  • Interactive flow

Primitive Operations

  • Using the primitives
  • Updating priors
  • Sample models
  • Fitting models
  • Pruning models
  • Visualise a model
  • Diverse models
  • Updating a QLattice
  • Validate data
  • Semantic types

Advanced

  • Converting a model to SymPy
  • Logging in Feyn
  • Setting themes
  • Saving a graph as an image
  • Using the query language
  • Estimating priors
  • Filtering models
  • Model parameters
  • Model complexity

Privacy & Commercial

  • Privacy
  • Community edition
  • Commercial use
  • Transition to Feyn 3.0

Residuals plot

by: Kevin Broløs & Chris Cave
(Feyn version 3.0 or newer)


Aside from the training metrics, Feyn offers a range of tools to help you evaluate your Model.

One of the basic diagnostics we can do with a regression Model is to plot the residuals (y_true - y_pred, the difference between the prediction and the truth).

This can help analyse whether errors are normally distributed or not. If they have an unusual distribution then it points towards biases in the Model. If they appear to be randomly scattered then this is a positive sign that the Model is unbiased.

Example

As sample data we are going for the Diabetes dataset made available by scikit-learn. Below we import data, prepare it and find a good Model from a QLattice:

import feyn

from sklearn.datasets import load_diabetes
import pandas as pd

from feyn.tools import split

# Load diabetes dataset into a pandas dataframe
dataset = load_diabetes()
df_diabetes = pd.DataFrame(dataset.data, columns=dataset.feature_names)
df_diabetes['response'] = dataset.target

# Train/test split
train, test = split(df_diabetes, ratio=[0.6, 0.4])

# Instantiate a QLattice
ql = feyn.QLattice()

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

Plotting the residuals

best.plot_residuals(data=train)

Residuals plot showing the distribution of the errors

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_residuals(data=train, filename="feyn-plot")

If the extension is not specified then it is saved as a png file.

Location in Feyn

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

from feyn.plots import plot_residuals

y_true = train['response']
y_pred = best.predict(train)

plot_residuals(y_true, y_pred)
← Regression plotROC curve →
  • Example
    • Plotting the residuals
    • Saving the plot
  • Location in Feyn

Subscribe to get news about Feyn and the QLattice.

You can opt out at any time, and you can read our privacy policy here.

Copyright © 2024 Abzu.ai - Feyn license: CC BY-NC-ND 4.0
Feyn®, QGraph®, and the QLattice® are registered trademarks of Abzu®