Intan Putri Setyowati

This project is a machine learning model that classifies images of Tom and Jerry. Built using Python in a Jupyter Notebook, it uses image preprocessing and a convolutional neural network (CNN) to distinguish between the two cartoon characters. The model is trained and evaluated using TensorFlow/Keras.

GITHUB REPOSITORIES ➔

THE RESULT

The model achieved a validation accuracy of 95.90%, indicating strong performance in correctly classifying Tom and Jerry images. The validation loss of 0.1443 is low, suggesting the model generalizes well without significant overfitting.

From the accuracy and loss plots, we can observe that training and validation accuracy both steadily increased, while the loss consistently decreased. This means the model learned effectively and maintained stable performance across epochs.

The model achieved a validation accuracy of 95.90%, indicating strong performance in correctly classifying Tom and Jerry images. The validation loss of 0.1443 is low, suggesting the model generalizes well without significant overfitting.

From the accuracy and loss plots, we can observe that training and validation accuracy both steadily increased, while the loss consistently decreased. This means the model learned effectively and maintained stable performance across epochs.

The model achieved a validation accuracy of 95.90%, indicating strong performance in correctly classifying Tom and Jerry images. The validation loss of 0.1443 is low, suggesting the model generalizes well without significant overfitting.

From the accuracy and loss plots, we can observe that training and validation accuracy both steadily increased, while the loss consistently decreased. This means the model learned effectively and maintained stable performance across epochs.

From the confusion matrix:

194 Jerry images were correctly classified,

while 10 were incorrectly predicted as Tom.

297 Tom images were correctly classified,

while 11 were incorrectly predicted as Jerry.

These results indicate that the model is highly accurate in distinguishing between the two classes, with very few misclassifications. The low number of errors on both sides suggests the model has learned the distinguishing features of Tom and Jerry well, without showing significant bias toward one class. The performance is balanced, with both classes achieving high precision and recall, reinforcing that the model is reliable and generalizes effectively on unseen data.

From the confusion matrix:

194 Jerry images were correctly classified,

while 10 were incorrectly predicted as Tom.

297 Tom images were correctly classified,

while 11 were incorrectly predicted as Jerry.

These results indicate that the model is highly accurate in distinguishing between the two classes, with very few misclassifications. The low number of errors on both sides suggests the model has learned the distinguishing features of Tom and Jerry well, without showing significant bias toward one class. The performance is balanced, with both classes achieving high precision and recall, reinforcing that the model is reliable and generalizes effectively on unseen data.

From the confusion matrix:

194 Jerry images were correctly classified, while 10 were incorrectly predicted as Tom.

297 Tom images were correctly classified, while 11 were incorrectly predicted as Jerry.

These results indicate that the model is highly accurate in distinguishing between the two classes, with very few misclassifications. The low number of errors on both sides suggests the model has learned the distinguishing features of Tom and Jerry well, without showing significant bias toward one class. The performance is balanced, with both classes achieving high precision and recall, reinforcing that the model is reliable and generalizes effectively on unseen data.

VISUALIZATION OF PREDICTION