Keras Functional API Models

The Keras Functional API provides a way to create models in a more flexible and modular way. It allows you to define the layers of your model as a sequence or a graph of layers, which can be connected in various ways.

Key Features

Modularity: Define layers as building blocks and connect them in various configurations.
Flexibility: Suitable for complex models where you need to define custom layers or connections.
Reusability: Layers can be reused across different models.

Basic Usage

Here's a simple example of how to use the Functional API to create a model:

from keras.models import Model
from keras.layers import Input, Dense

# Define the input
input_tensor = Input(shape=(32,))

# Define the first layer
x = Dense(64, activation='relu')(input_tensor)

# Define the second layer
output_tensor = Dense(10, activation='softmax')(x)

# Create the model
model = Model(inputs=input_tensor, outputs=output_tensor)

Layers

Keras provides a wide range of layers that can be used in the Functional API. Here are some commonly used layers:

Dense: Fully connected layer.
Conv2D: Convolutional layer for 2D data (e.g., images).
MaxPooling2D: Pooling layer for 2D data.
Dropout: Dropout layer for regularization.

Model Compilation

After defining the model, you need to compile it with an optimizer, a loss function, and metrics:

model.compile(optimizer='adam',
              loss='categorical_crossentropy',
              metrics=['accuracy'])

Model Training

You can train the model using the fit method:

model.fit(x_train, y_train, epochs=10, batch_size=32)