add pytest 'backwards' function

tests/test_utils_memory.py

@@ -1,6 +1,6 @@
 import pytest
 import pandas as pd
-from source.utils.utils_memory import update_period2farm_and_farm2period_train, update_period2farm_and_farm2period_test
+from source.utils.utils_memory import update_period2farm_and_farm2period_train, update_period2farm_and_farm2period_test, backwards
 
 # Test case: valid DataFrame
 def test_update_period2farm_and_farm2period_train_output(sample_df):
@@ -57,3 +57,22 @@ def test_update_period2farm_and_farm2period_test_missing_columns(sample_df_missi
     # Call the function with a DataFrame missing required columns
     with pytest.raises(AssertionError, match="DataFrame df must contain columns 'periodId', 'farmId', and 'power_z'."):
         update_period2farm_and_farm2period_test(sample_df_missing_columns)
+
+# Test case: DataFrame output contain lags
+def test_backwards_output(sample_df_back_lags):
+    " Test if the lag features are created correctly. "
+    # Call the function with a DataFrame
+    df_lags = backwards(sample_df_back_lags, ['A', 'B'], [1, 2], 2)
+    # Assert that the lag features are created correctly
+    print(df_lags.columns.tolist())
+    assert df_lags.columns.tolist() == ['A', 'B', 'A_lag_-1', 'A_lag_-2', 'B_lag_-1', 'B_lag_-2'], "Lag features are not created correctly."
+
+# Test case: check that no missing values are present
+def test_backwards_no_missing_values(sample_df_back_lags):
+    " Test if the lag features do not contain missing values. "
+    # Call the function with a DataFrame
+    df_lags = backwards(sample_df_back_lags, ['A'], [1, 2], 2)
+    # Assert that there are no missing values in the lag features
+    print(df_lags)
+    print(df_lags.isnull().sum())
+    assert df_lags.isnull().sum().sum() == 0, "Lag features should not contain missing values."