Hands-On Machine Learning for Algorithmic Trading

Running GridSearchCV to tune the neural network architecture

We split the data into a training set for cross-validation and a holdout test set using stratified sampling, as the classes are slightly unbalanced, as follows:

X_train, X_test, y_train, y_test = train_test_split(features, label,                                                   test_size=.1,                                                   random_state=42,                                                   shuffle=True,                                                   stratify=data.label)

Now we just need to define our Keras classifier using the make_model function, set stratified cross-validation, and define the parameters that we would like to explore, as follows:

clf = KerasClassifier(make_model, epochs=10, batch_size=32)cv = StratifiedKFold(n_splits=5, shuffle=True)param_grid = {'dense_layers': [[64], [64, 64], [96, 96], [128, 128]],              'optimizer': ['RMSprop', 'Adam'],