def test_expanding_sum_single_var(df_time):
expected_results = {
"ambient_temp_expanding_sum": [
np.nan,
31.31,
62.82,
94.97,
127.36,
159.98,
192.48,
225.00,
257.68,
291.44,
325.57,
359.65,
393.35,
427.24,
461.28,
],
}
expected_df = df_time.copy()
expected_df["ambient_temp_expanding_sum"] = expected_results[
"ambient_temp_expanding_sum"]
transformer = ExpandingWindowFeatures(variables=["ambient_temp"],
functions="sum")
df_tr = transformer.fit_transform(df_time)
assert_frame_equal(df_tr, expected_df)
def test_expanding_sum_and_mean_single_var(df_time):
expected_results = {
"ambient_temp_expanding_sum": [
np.nan,
31.31,
62.82,
94.97,
127.36,
159.98,
192.48,
225.00,
257.68,
291.44,
325.57,
359.65,
393.35,
427.24,
461.28,
],
"ambient_temp_expanding_mean": [
np.nan,
31.3100,
31.4100,
31.6567,
31.8400,
31.9960,
32.0800,
32.1429,
32.2100,
32.3822,
32.5570,
32.6955,
32.7792,
32.8646,
32.9486,
],
}
expected_df = df_time.copy()
expected_df["ambient_temp_expanding_sum"] = expected_results[
"ambient_temp_expanding_sum"]
expected_df["ambient_temp_expanding_mean"] = expected_results[
"ambient_temp_expanding_mean"]
transformer = ExpandingWindowFeatures(variables=["ambient_temp"],
functions=["sum", "mean"])
df_tr = transformer.fit_transform(df_time)
assert_frame_equal(df_tr, expected_df)
def test_expanding_sum_multiple_vars(df_time):
expected_results = {
"ambient_temp_expanding_sum": [
np.nan,
31.31,
62.82,
94.97,
127.36,
159.98,
192.48,
225.00,
257.68,
291.44,
325.57,
359.65,
393.35,
427.24,
461.28,
],
"irradiation_expanding_sum": [
np.nan,
0.51,
1.3,
1.95,
2.71,
3.13,
3.62,
4.19,
4.75,
5.49,
6.38,
6.85,
7.39,
7.79,
8.24,
],
}
expected_df = df_time.copy()
expected_df["ambient_temp_expanding_sum"] = expected_results[
"ambient_temp_expanding_sum"]
expected_df["irradiation_expanding_sum"] = expected_results[
"irradiation_expanding_sum"]
transformer = ExpandingWindowFeatures(
variables=["ambient_temp", "irradiation"], functions="sum")
df_tr = transformer.fit_transform(df_time)
assert_frame_equal(df_tr, expected_df)
def test_get_feature_names_out_raises_when_input_features_is_string(df_time):
tr = ExpandingWindowFeatures(functions=["mean", "sum"])
tr.fit(df_time)
with pytest.raises(ValueError):
# get error when user does not pass a list
tr.get_feature_names_out(input_features="ambient_temp")
def test_get_feature_names_out_raises_when_input_features_not_transformed(
df_time):
tr = ExpandingWindowFeatures(functions=["mean", "sum"])
tr.fit(df_time)
with pytest.raises(ValueError):
# assert error when uses passes features that were not transformed
tr.get_feature_names_out(input_features=["color"])
def test_sort_index(df_time):
# Shuffle dataframe
Xs = df_time.sample(frac=1)
transformer = ExpandingWindowFeatures(sort_index=False)
df_tr = transformer.fit_transform(Xs)
assert_frame_equal(df_tr[transformer.variables_],
Xs[transformer.variables_])
transformer = ExpandingWindowFeatures(sort_index=True)
df_tr = transformer.fit_transform(Xs)
assert_frame_equal(df_tr[transformer.variables_],
Xs[transformer.variables_].sort_index())
def test_get_feature_names_out_single_variable_and_single_function(df_time):
# input features
original_features = ["ambient_temp", "module_temp", "irradiation", "color"]
tr = ExpandingWindowFeatures(variables="ambient_temp", functions="sum")
tr.fit(df_time)
# expected
output = [
"ambient_temp_expanding_sum",
]
assert tr.get_feature_names_out(
input_features=None) == original_features + output
assert tr.get_feature_names_out(input_features=["ambient_temp"]) == output
def test_get_feature_names_out_multiple_variables_and_functions(df_time):
# input features
input_features = ["ambient_temp", "module_temp", "irradiation"]
original_features = ["ambient_temp", "module_temp", "irradiation", "color"]
tr = ExpandingWindowFeatures(functions=["mean", "sum"])
tr.fit(df_time)
# expected
output = [
"ambient_temp_expanding_mean",
"ambient_temp_expanding_sum",
"module_temp_expanding_mean",
"module_temp_expanding_sum",
"irradiation_expanding_mean",
"irradiation_expanding_sum",
]
assert tr.get_feature_names_out(
input_features=None) == original_features + output
assert tr.get_feature_names_out(input_features=input_features) == output
assert tr.get_feature_names_out(
input_features=input_features[0:2]) == output[0:4]
assert tr.get_feature_names_out(
input_features=[input_features[0]]) == output[0:2]
import pandas as pd
import pytest
from sklearn.base import clone
from sklearn.utils.estimator_checks import check_estimator
from tests.estimator_checks.estimator_checks import check_feature_engine_estimator
from feature_engine.timeseries.forecasting import (
ExpandingWindowFeatures,
LagFeatures,
WindowFeatures,
)
_estimators = [
LagFeatures(missing_values="ignore"),
WindowFeatures(missing_values="ignore"),
ExpandingWindowFeatures(missing_values="ignore"),
]
@pytest.mark.parametrize("estimator", _estimators)
def test_check_estimator_from_sklearn(estimator):
return check_estimator(estimator)
@pytest.mark.parametrize("estimator", _estimators)
def test_check_estimator_from_feature_engine(estimator):
return check_feature_engine_estimator(estimator)
@pytest.mark.parametrize("estimator", _estimators)
def test_error_when_not_unique_values_in_index(df_time, estimator):
def test_expanding_window_raises_when_periods_negative():
with pytest.raises(
ValueError,
match="periods must be a non-negative integer. Got -1 instead."):
ExpandingWindowFeatures(periods=-1)
def test_permitted_param_periods(_periods):
transformer = ExpandingWindowFeatures(periods=_periods)
assert transformer.periods == _periods
def test_expanding_sum_and_mean_multiple_vars(df_time):
expected_results = {
"ambient_temp_expanding_sum": [
np.nan,
31.31,
62.82,
94.97,
127.36,
159.98,
192.48,
225.00,
257.68,
291.44,
325.57,
359.65,
393.35,
427.24,
461.28,
],
"ambient_temp_expanding_mean": [
np.nan,
31.3100,
31.4100,
31.6567,
31.8400,
31.9960,
32.0800,
32.1429,
32.2100,
32.3822,
32.5570,
32.6955,
32.7792,
32.8646,
32.9486,
],
"irradiation_expanding_sum": [
np.nan,
0.51,
1.3,
1.95,
2.71,
3.13,
3.62,
4.19,
4.75,
5.49,
6.38,
6.85,
7.39,
7.79,
8.24,
],
"irradiation_expanding_mean": [
np.nan,
0.51000,
0.65000,
0.65000,
0.67750,
0.62600,
0.60333,
0.59857,
0.59375,
0.61000,
0.63800,
0.62273,
0.61583,
0.59923,
0.58857,
],
}
expected_df = df_time.copy()
expected_df["ambient_temp_expanding_sum"] = expected_results[
"ambient_temp_expanding_sum"]
expected_df["ambient_temp_expanding_mean"] = expected_results[
"ambient_temp_expanding_mean"]
expected_df["irradiation_expanding_sum"] = expected_results[
"irradiation_expanding_sum"]
expected_df["irradiation_expanding_mean"] = expected_results[
"irradiation_expanding_mean"]
transformer = ExpandingWindowFeatures(
variables=["ambient_temp", "irradiation"], functions=["sum", "mean"])
df_tr = transformer.fit_transform(df_time)
assert_frame_equal(df_tr, expected_df)