. This allows analysts to quickly identify which classes the model is struggling with by visualizing key metrics: PyPI +1 Precision: How many of the positive predictions were actually correct. Recall: How many of the actual positive cases the model managed to find. F1 Score: The harmonic mean of precision and recall, providing a single score for overall balance. Support: The number of actual occurrences of each class in the dataset. Key Benefits for Analysts Visual Intuition: Instead of scanning a table of raw numbers, the heatmap helps detect Type I (false positive) and Type II (false negative) errors at a glance. Model Comparison: It allows for fast visual comparison between different algorithms or hyperparameter settings. Pipeline Integration: Like other Yellowbrick visualizers, it acts as a Scikit-Learn "Visualizer" object, meaning it can be integrated directly into your existing machine learning pipeline. PyPI +3 How to Generate a Report To create a Classification Report, you typically follow these steps using the
vc = ValidationCurve(LogisticRegression(), param_name="C", param_range=np.logspace(-4, 1, 6)) vc.fit(X, y) vc.show() # Find C where validation score peaks yellowbrick analyst tool
Show an AlphaSelection plot for Lasso/Ridge regression. F1 Score: The harmonic mean of precision and
from yellowbrick.classifier import ConfusionMatrix from sklearn.ensemble import RandomForestClassifier # 1. Initialize the visualizer model = RandomForestClassifier() visualizer = ConfusionMatrix(model) # 2. Fit and Score visualizer.fit(X_train, y_train) visualizer.score(X_test, y_test) # 3. Save as a publication-ready file (PNG, PDF, or SVG) visualizer.show(outpath="confusion_matrix.pdf") Use code with caution. Copied to clipboard Model Comparison: It allows for fast visual comparison
In ~10 lines of code, you’ve performed learning curve analysis, hyperparameter tuning, and multi-class ROC evaluation. That’s 2 hours of manual plotting compressed into 5 minutes.
Yellowbrick also handles unsupervised learning through Clustering Visualizers. Determining the number of clusters in K-Means is often a guessing game, but Yellowbrick’s Elbow Method and Silhouette Visualizers provide a mathematical and visual justification for choosing the optimal "K." It takes the guesswork out of clustering by showing how well-defined the boundaries are between groups.