Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
37 views4 pages

Performance Metrics ML

The document outlines essential performance metrics for evaluating machine learning models, distinguishing between classification and regression metrics. Key classification metrics include accuracy, precision, recall, F1 score, ROC-AUC, specificity, and log loss, while regression metrics cover mean absolute error, mean squared error, root mean squared error, and R² score. Each metric serves specific purposes, such as measuring correctness, relevance, and variance capture in model predictions.

Uploaded by

vimalrajk.a.12.b3.2021.22
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
37 views4 pages

Performance Metrics ML

The document outlines essential performance metrics for evaluating machine learning models, distinguishing between classification and regression metrics. Key classification metrics include accuracy, precision, recall, F1 score, ROC-AUC, specificity, and log loss, while regression metrics cover mean absolute error, mean squared error, root mean squared error, and R² score. Each metric serves specific purposes, such as measuring correctness, relevance, and variance capture in model predictions.

Uploaded by

vimalrajk.a.12.b3.2021.22
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Performance Metrics for Evaluation of Machine Learning Models

Introduction

Performance metrics are essential tools in machine learning to evaluate the effectiveness and accuracy of a

model.

Different metrics apply to different types of machine learning tasks such as classification and regression.

Classification Metrics

I. Classification Metrics

Used when the output variable is categorical (e.g., Yes/No, Spam/Not Spam). A confusion matrix is the

foundation for many classification metrics.

Confusion Matrix:

Predicted Yes Predicted No

Actual Yes True Positive False Negative

Actual No False Positive True Negative

1. Accuracy: (TP + TN) / (TP + TN + FP + FN)

- Measures overall correctness.

- Good for balanced datasets.


Performance Metrics for Evaluation of Machine Learning Models

2. Precision: TP / (TP + FP)

- Measures how many selected items are relevant.

- Important in spam detection.

3. Recall (Sensitivity): TP / (TP + FN)

- Measures how many relevant items are selected.

- Important in medical diagnoses.

4. F1 Score: 2 * (Precision * Recall) / (Precision + Recall)

- Harmonic mean of precision and recall.

- Balances false positives and false negatives.

5. ROC-AUC: Shows the trade-off between TPR and FPR.

- AUC close to 1 means a better model.

6. Specificity: TN / (TN + FP)

- Measures the proportion of actual negatives correctly identified.

7. Log Loss: Penalizes false classifications with confidence.

- Lower value indicates better performance.

Regression Metrics

II. Regression Metrics


Performance Metrics for Evaluation of Machine Learning Models

Used when the output variable is continuous (e.g., predicting price, temperature).

1. Mean Absolute Error (MAE): Average of absolute errors.

- Easy to interpret.

2. Mean Squared Error (MSE): Average of squared errors.

- Penalizes large errors.

3. Root Mean Squared Error (RMSE): Square root of MSE.

- Same units as target.

4. R² Score: 1 - (Sum of Squares Error / Total Sum of Squares)

- Explains how well the model captures variance.

Summary

Summary Table:

| Metric | Type | Good For |

|------------|----------------|----------------------------------|

| Accuracy | Classification | Balanced class distributions |

| Precision | Classification | Avoiding false positives |

| Recall | Classification | Avoiding false negatives |

| F1 Score | Classification | When precision & recall matter |


Performance Metrics for Evaluation of Machine Learning Models

| ROC-AUC | Classification | Evaluating classifiers |

| MAE | Regression | Simple average error |

| MSE/RMSE | Regression | Penalizing larger errors |

| R² Score | Regression | Explained variance |

You might also like