def test_evaluate_regression_optimiser():
"""Test evaluate method of Optimiser class for regression."""
reader = Reader(sep=",")
dict = reader.train_test_split(Lpath=[
"data_for_tests/train_regression.csv",
"data_for_tests/test_regression.csv"
],
target_name="SalePrice")
drift_thresholder = Drift_thresholder()
drift_thresholder = drift_thresholder.fit_transform(dict)
mape = make_scorer(lambda y_true, y_pred: 100 * np.sum(
np.abs(y_true - y_pred) / y_true) / len(y_true),
greater_is_better=False,
needs_proba=False)
with pytest.warns(UserWarning) as record:
opt = Optimiser(scoring=mape, n_folds=3)
assert len(record) == 1
score = opt.evaluate(None, dict)
assert -np.Inf <= score
with pytest.warns(UserWarning) as record:
opt = Optimiser(scoring=None, n_folds=3)
assert len(record) == 1
score = opt.evaluate(None, dict)
assert -np.Inf <= score
with pytest.warns(UserWarning) as record:
opt = Optimiser(scoring="wrong_scoring", n_folds=3)
assert len(record) == 1
with pytest.warns(UserWarning) as record:
score = opt.evaluate(None, dict)
assert -np.Inf <= score
def test_fit_predict_predictor_regression(mock_show):
"""Test fit_predict method of Predictor class for regression."""
rd = Reader(sep=',')
dt = rd.train_test_split(Lpath=[
"data_for_tests/train_regression.csv",
"data_for_tests/test_regression.csv"
],
target_name="SalePrice")
drift_thresholder = Drift_thresholder()
df = drift_thresholder.fit_transform(dt)
mape = make_scorer(lambda y_true, y_pred: 100 * np.sum(
np.abs(y_true - y_pred) / y_true) / len(y_true),
greater_is_better=False,
needs_proba=False)
opt = Optimiser(scoring=mape, n_folds=3)
opt.evaluate(None, df)
space = {
'ne__numerical_strategy': {
"search": "choice",
"space": [0]
},
'ce__strategy': {
"search":
"choice",
"space":
["label_encoding", "random_projection", "entity_embedding"]
},
'fs__threshold': {
"search": "uniform",
"space": [0.01, 0.3]
},
'est__max_depth': {
"search": "choice",
"space": [3, 4, 5, 6, 7]
}
}
best = opt.optimise(space, df, 1)
prd = Predictor(verbose=True)
prd.fit_predict(best, df)
pred_df = pd.read_csv("save/SalePrice_predictions.csv")
assert np.all(
list(pred_df.columns) == ['Unnamed: 0', 'SalePrice_predicted'])
assert np.shape(pred_df) == (1459, 2)
# Reading and cleaning all files
# Declare a reader for csv files
rd = Reader(sep=',')
# Return a dictionnary containing three entries
# dict["train"] contains training samples withtout target columns
# dict["test"] contains testing elements withtout target columns
# dict["target"] contains target columns for training samples.
dict = rd.train_test_split(paths, target_name)
dft = Drift_thresholder()
dict = dft.fit_transform(dict)
# Tuning
mape = make_scorer(lambda y_true, y_pred: 100 * np.sum(
np.abs(y_true - y_pred) / y_true) / len(y_true),
greater_is_better=False,
needs_proba=False)
# Declare an optimiser. You can declare your own score
# as presented here or use one in
# {"mean_absolute_error", "mean_squared_error","median_absolute_error","r2"}
opt = Optimiser(scoring=mape, n_folds=3)
opt.evaluate(None, dict)
# Space of hyperparameters
# The keys must respect the following syntax : "enc__param".
# "enc" = "ne" for na encoder
# "enc" = "ce" for categorical encoder
# "enc" = "fs" for feature selector [OPTIONAL]
# "enc" = "stck"+str(i) to add layer n°i of meta-features [OPTIONAL]
# "enc" = "est" for the final estimator
# "param" : a correct associated parameter for each step.