Model Evaluation

Step 9.1: Plot Training Curves

• Metrics Visualization: Plot the loss, binary accuracy, and subtype accuracy for both training and validation sets across all epochs.

  • Purpose: Helps understand how the model is learning and whether it is overfitting or underfitting.

    

Step 9.2: Evaluate on Test Set

• Evaluation Function: The evaluate_model() function is used to evaluate the model's performance on the test dataset.

  • Metrics: Generates a classification report and confusion matrix for both binary and subtype classification.

  • Purpose: Provides a detailed breakdown of model performance, including precision, recall, and F1-score for each class.

    

Step 9.3: Display Predictions

• Display Predictions: The show_predictions() function is used to display model predictions for a few test samples.

  • Visualization: Shows the true and predicted labels for each sample.

  • Purpose: Visually verify the model's predictions and understand its performance on individual images.