'1200rpm',
'variable_rpm',
],
'fit_modes': [
'per_feature',
'per_sample',
],
}
run_configs = {
'data_set': config['data_sets'],
'model_function': config['model_functions'],
'scaling': config['scalings'],
'fit_mode': config['fit_modes'],
}
experiment = Experiment()
experiment.log('config/config', config)
experiment.log('config/run_configs', run_configs)
def run_callable(run_config: dict):
def modifier(x):
return x[x.rpm > 0]
def pre_processing(data_frame):
if run_config['scaling'] == 'min_max':
samples = Scaler(MinMaxScaler, fit_mode=run_config['fit_mode']).fit_transform(data_frame.to_numpy())
else:
samples = data_frame.to_numpy()
return build_samples(samples, target_sample_length=config['input_size'], target_dimensions=3)
'800rpm_gradual',
# '1200rpm',
# 'variable_rpm'
],
'fit_modes': [
'per_feature',
# 'per_sample'
],
}
run_configs = {
'data_set': config['data_sets'],
'fit_mode': config['fit_modes'],
'scaling': config['scalings'],
}
experiment = Experiment(auto_datetime_directory=False)
experiment.log('config/config', config)
experiment.log('config/run_configs', run_configs)
model = create_feed_forward_autoencoder(
input_dimension=config['input_size'],
encoding_dimension=config['encoding_size'],
hidden_layer_activations=config['hidden_layer_activations'],
output_layer_activation=config['output_layer_activation'],
loss=config['loss'],
)
def run_callable(run_config: dict):
experiment.print('Loading data')
bearing_dataset = load_data(run_config['data_set'])
'a2e.evaluation.reconstruction_error_cost',
],
}
config_space = create_config_space(
min_dropout_rate_input=.2,
max_dropout_rate_input=.8,
min_dropout_rate_hidden_layers=.2,
max_dropout_rate_hidden_layers=.8,
min_dropout_rate_output=.2,
max_dropout_rate_output=.8,
)
if __name__ == '__main__':
experiment = Experiment(auto_datetime_directory=True)
experiment.log('config/config', config)
experiment.log('config/run_configs', run_configs)
experiment.log('config/config_space', str(config_space))
def run_callable(run_config: dict):
experiment.print('Loading data')
bearing_dataset = load_data(run_config['data_set'])
x_train = bearing_dataset.train(column=config['data_column'], as_numpy=True)
experiment.print('Initializing optimizer')
experiment.print(f'max_iterations = {config["max_iterations"]}')
optimizer = create_optimizer(
run_config['optimizer'],
config_space=config_space,
model=KerasModel(