def test_custom_metric(self):
automl_experiment = AutoML()
automl_settings = {
"time_budget": 10,
'eval_method': 'holdout',
"metric": custom_metric,
"task": 'classification',
"log_file_name": "test/iris_custom.log",
"log_training_metric": True,
'log_type': 'all',
"model_history": True
}
X_train, y_train = load_iris(return_X_y=True)
automl_experiment.fit(X_train=X_train,
y_train=y_train,
**automl_settings)
print(automl_experiment.classes_)
print(automl_experiment.predict_proba(X_train))
print(automl_experiment.model)
print(automl_experiment.config_history)
print(automl_experiment.model_history)
print(automl_experiment.best_iteration)
print(automl_experiment.best_estimator)
automl_experiment = AutoML()
estimator = automl_experiment.get_estimator_from_log(
automl_settings["log_file_name"], record_id=0, objective='multi')
print(estimator)
time_history, best_valid_loss_history, valid_loss_history, \
config_history, train_loss_history = get_output_from_log(
filename=automl_settings['log_file_name'], time_budget=6)
print(train_loss_history)
def test_custom_metric(self):
df, y = load_iris(return_X_y=True, as_frame=True)
df["label"] = y
automl_experiment = AutoML()
automl_settings = {
"dataframe": df,
"label": "label",
"time_budget": 5,
"eval_method": "cv",
"metric": custom_metric,
"task": "classification",
"log_file_name": "test/iris_custom.log",
"log_training_metric": True,
"log_type": "all",
"n_jobs": 1,
"model_history": True,
"sample_weight": np.ones(len(y)),
"pred_time_limit": 1e-5,
"ensemble": True,
}
automl_experiment.fit(**automl_settings)
print(automl_experiment.classes_)
print(automl_experiment.model)
print(automl_experiment.config_history)
print(automl_experiment.best_model_for_estimator("rf"))
print(automl_experiment.best_iteration)
print(automl_experiment.best_estimator)
automl_experiment = AutoML()
estimator = automl_experiment.get_estimator_from_log(
automl_settings["log_file_name"], record_id=0, task="multi"
)
print(estimator)
(
time_history,
best_valid_loss_history,
valid_loss_history,
config_history,
metric_history,
) = get_output_from_log(
filename=automl_settings["log_file_name"], time_budget=6
)
print(metric_history)
try:
import ray
df = ray.put(df)
automl_settings["dataframe"] = df
automl_settings["use_ray"] = True
automl_experiment.fit(**automl_settings)
except ImportError:
pass
