Neural Network Tutorials - Herong's Tutorial Examples

∟Performance Evaluation Metrics

∟PCC (Pearson Correlation Coefficient)

This section describes PCC (Pearson Correlation Coefficient) as a metric to evaluate the performance of a continuous prediction model.

What Is PCC (Pearson Correlation Coefficient)? - PCC (Pearson Correlation Coefficient) is a commonly used metric to evaluate the performance of a continuous prediction model. PCC measures how well predicted values are correlated to actual values.

Given a prediction model and a set of test samples, the PCC of the model on the test set is defined below:

where:

PCC can also be expressed as:

where:

Interpretations of PCC:

• If PCC = 1, a linear equation describes the relationship between predicted values and actual values perfectly, with all data points lying on a line. This indicates a good prediction.
• If PCC = 0, there is no linear dependency between predicted values and actual values. This indicates a poor prediction. Actually, a random prediction will result PCC = 0.
• If PCC = -1, all data points lying on a line in a reverse order. This indicates that something went wrong in the prediction.

Table of Contents

 About This Book

 Deep Playground for Classical Neural Networks

 Building Neural Networks with Python

 Simple Example of Neural Networks

 TensorFlow - Machine Learning Platform

 PyTorch - Machine Learning Platform

 Gradio - ML Demo Platform

 CNN (Convolutional Neural Network)

 RNN (Recurrent Neural Network)

 GNN (Graph Neural Network)

 GAN (Generative Adversarial Network)

►Performance Evaluation Metrics

 MSE (Mean Squared Error)

 CI (Concordance Index)

►PCC (Pearson Correlation Coefficient)

 References

 Full Version in PDF/EPUB

PCC (Pearson Correlation Coefficient) - Updated in 2023, by Herong Yang