Implementation of a custom TensorFlow/Keras model by subclassing tf.keras.Model, including custom layers, a user-defined loss function, and full control over the forward pass. Demonstrates dataset loading, preprocessing, and training on MNIST using the Model Subclassing API for maximum flexibility
pythonapideep-learninginteractivetensorflownumpykerasclassificationmodel-subclassinglowlevelapicustomsubclassmodellow-level-api-tensorflowcustom-layar--model
-
Updated
Aug 17, 2025 - Jupyter Notebook