TensorFlow & Keras · Module 3 of 10

Building models with Keras

Create networks with Sequential and Functional APIs.

📖 Key ideas

Sequential: keras.Sequential([layers.Dense(64, activation="relu"), layers.Dense(10, activation="softmax")]).
Functional API builds models with multiple inputs/outputs.
Subclassing keras.Model gives full flexibility.
model.summary() shows layers and parameters.
Common layers: Dense, Conv2D, MaxPooling2D, Dropout, LSTM, Embedding.

💡 Real-life example: Sequential models are a straight train; Functional models are a railway network with junctions.

SequentialLinear stack of layers

Functional APIGraph of layers

DenseFully connected layer

model.summary()Shows architecture