Last Updated: November 21, 2025
Keras
High-level neural networks API
Core Components
| Item | Description |
|---|---|
Sequential
|
Linear stack of layers |
Functional API
|
Complex model architectures |
Layer
|
Building block of models |
Model
|
Training and inference |
Callback
|
Training hooks |
Optimizer
|
Training algorithms |
Model Example
from tensorflow import keras
from tensorflow.keras import layers
# Sequential model
model = keras.Sequential([
layers.Dense(128, activation='relu', input_shape=(784,)),
layers.Dropout(0.2),
layers.Dense(10, activation='softmax')
])
model.compile(
optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy']
)
# Train
history = model.fit(
x_train, y_train,
epochs=10,
validation_split=0.2,
batch_size=32
)
# Evaluate
model.evaluate(x_test, y_test)Common Layers
| Item | Description |
|---|---|
Dense
|
Fully connected layer |
Conv2D
|
2D convolution |
MaxPooling2D
|
Max pooling |
Dropout
|
Regularization |
LSTM
|
Recurrent layer |
BatchNormalization
|
Normalize activations |
Best Practices
- Use functional API for complex architectures
- Add callbacks for early stopping
- Use validation split to monitor overfitting
- Save best model with ModelCheckpoint
💡 Pro Tips
Quick Reference
Keras is now integrated into TensorFlow 2.0+