keras_tuner_randomsearch_hyperparameter_tuning_units_learning_rate.py

1import keras
2from keras import layers
3import keras_tuner
4
5def build_model(hp):
6    model = keras.Sequential()
7    model.add(layers.Flatten())
8    model.add(
9        layers.Dense(
10            # Define the hyperparameter
11            units=hp.Int("units", min_value=32, max_value=512, step=32),
12            activation="relu",
13        )
14    )
15    model.add(layers.Dense(10, activation="softmax"))
16    model.compile(
17        optimizer=keras.optimizers.Adam(
18            hp.Choice("learning_rate", values=[1e-2, 1e-3, 1e-4])
19        ),
20        loss="sparse_categorical_crossentropy",
21        metrics=["accuracy"],
22    )
23    return model
24
25# Initialize the tuner
26tuner = keras_tuner.RandomSearch(
27    build_model,
28    objective="val_accuracy",
29    max_trials=3,
30    executions_per_trial=2,
31    overwrite=True,
32    directory="my_dir",
33    project_name="helloworld",
34)
35
36# Prepare data
37(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()
38x_train = x_train.astype("float32") / 255
39x_test = x_test.astype("float32") / 255
40
41# Start the search
42tuner.search(x_train, y_train, epochs=2, validation_data=(x_test, y_test))
43
44# Get the best model
45best_model = tuner.get_best_models(num_models=1)[0]