n_init=[10],
epsilon=[1e-05],
patience=[10],
)
params_architecture: Mapping[str, Sequence[Any]] = dict(
size_hidden_layers=[(2, )],
activation=[tanh_classification],
activation_hidden=[relu],
eta=[0.1, 0.01, 0.001],
alpha=[0, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],
alambd=[0, 0.0001, 0.001, 0.01],
eta_decay=[0],
)
cv_params: Mapping[str, Any] = dict(
cv=3,
error_calculator=ErrorCalculator.MSE,
to_shuffle=True,
)
grid_search_results = grid_search(
train_data,
params_nn=params_nn,
params_architecture=params_architecture,
cv_params=cv_params,
n_jobs=8,
)
write_on_file(grid_search_results, 'results/monk1-mse-low_eta')
learning_algorithm=[batch],
epochs_limit=[1000],
n_init=[1],
epsilon=[1e-05],
patience=[10],
)
params_architecture: Mapping[str, Sequence[Any]] = dict(
size_hidden_layers=[(2, )],
activation=[tanh_classification],
activation_hidden=[relu],
eta=[0.1],
alpha=[0.7],
alambd=[0.0001],
eta_decay=[0],
)
validation_params: Mapping[str, Any] = dict(
validation_set=validation_set,
error_calculator=ErrorCalculator.MSE,
)
grid_search_results = grid_search(
train_set,
params_nn=params_nn,
params_architecture=params_architecture,
validation_params=validation_params,
n_jobs=8,
)
write_on_file(grid_search_results, 'results/monk1_choose_seed')