ADD: Data tests (#207)

choice_learn/datasets/base.py

@@ -645,7 +645,7 @@ def load_modecanada(
             if row.choice == 1:
                 named_choice[n_row - 1] = row.alt
 
-        canada_df["choice"] = named_choice
+        canada_df["named_choice"] = named_choice
 
     if as_frame:
         if split_features:
@@ -667,6 +667,9 @@ def load_modecanada(
                 items_features_by_choice,
                 choices,
             )
+        if choice_format == "items_id":
+            canada_df["choice"] = canada_df["named_choice"]
+            canada_df = canada_df.drop("named_choice", axis=1)
         return canada_df
 
     if split_features:
@@ -687,7 +690,7 @@ def load_modecanada(
                     cf.append(context_df.loc[context_df.alt == item][items_features].to_numpy()[0])
                     cav.append(1)
                 else:
-                    cf.append([0.0, 0.0, 0.0, 0.0])
+                    cf.append([0.0 for _ in range(len(items_features))])
                     cav.append(0)
             cif.append(cf)
             ci_av.append(cav)

tests/data/test_data.csv

@@ -0,0 +1,21 @@
+,srch_id,prop_id,site_id
+0,1,893,12
+1,1,10404,12
+2,1,21315,12
+3,1,27348,12
+4,1,29604,12
+5,1,30184,12
+6,1,44147,12
+7,1,50984,12
+8,1,53341,12
+9,1,56880,12
+10,1,59267,12
+11,1,59526,12
+12,1,68914,12
+13,1,74474,12
+14,1,81437,12
+15,1,85728,12
+16,1,88096,12
+17,1,88127,12
+18,1,88218,12
+19,1,89073,12

tests/unit_tests/datasets/test_expedia.py

@@ -0,0 +1,11 @@
+"""Unit testing for Expedia loader."""
+
+import pytest
+
+from choice_learn.datasets import load_expedia
+
+
+def test_raise_filenotfound():
+    """Test that error raised if no file exist."""
+    with pytest.raises(FileNotFoundError):
+        load_expedia()

tests/unit_tests/test_os_datasets.py

@@ -15,6 +15,7 @@
     load_tafeng,
     load_train,
 )
+from choice_learn.datasets.base import load_csv, load_gzip, slice_from_names
 
 
 def test_swissmetro_loader():
@@ -29,15 +30,73 @@ def test_swissmetro_loader():
     assert isinstance(swissmetro, ChoiceDataset)
 
 
+def test_swissmetro_long_format():
+    """Test loading the Swissmetro dataset in long format."""
+    swissmetro = load_swissmetro(as_frame=True, preprocessing="long_format")
+    assert isinstance(swissmetro, pd.DataFrame)
+    assert swissmetro.shape == (30474, 7)
+
+
+def test_swissmetro_tastenet():
+    """Test TasteNet preprocessing of dataset."""
+    _ = load_swissmetro(preprocessing="tastenet")
+
+
+def test_swissmetro_tutorial():
+    """Test tutorial preprocessing of dataset."""
+    _ = load_swissmetro(preprocessing="tutorial")
+
+
+def test_biogeme_nested_tutorial():
+    """Test biogeme_nested preprocessing of dataset."""
+    _ = load_swissmetro(preprocessing="biogeme_nested")
+
+
+def test_rumnet_tutorial():
+    """Test rumnet preprocessing of dataset."""
+    _ = load_swissmetro(preprocessing="rumnet")
+
+
 def test_modecanada_loader():
     """Test loading the Canada dataset."""
-    canada = load_modecanada(as_frame=True)
+    canada = load_modecanada(as_frame=True, choice_format="items_id")
     assert isinstance(canada, pd.DataFrame)
     assert canada.shape == (15520, 11)
 
     canada = load_modecanada()
     assert isinstance(canada, ChoiceDataset)
 
+    ca, na, da = load_modecanada(
+        as_frame=True,
+        add_items_one_hot=True,
+        add_is_public=True,
+        choice_format="items_id",
+        split_features=True,
+    )
+    assert ca.shape == (4324, 4)
+    assert na.shape == (15520, 11)
+    assert da.shape == (4324, 2)
+
+
+def test_modecanada_features_split():
+    """Test that features are split well."""
+    (
+        o,
+        ca,
+        na,
+        da,
+    ) = load_modecanada(add_items_one_hot=True, add_is_public=True, split_features=True)
+    assert o.shape == (4324, 3)
+    assert ca.shape == (4324, 4, 9)
+    assert na.shape == (4324, 4)
+    assert da.shape == (4324,)
+
+
+def test_modecanada_loader_2():
+    """Test loading the Canada dataset w/ preprocessing."""
+    canada = load_modecanada(preprocessing="tutorial", add_items_one_hot=True)
+    assert isinstance(canada, ChoiceDataset)
+
 
 def test_electricity_loader():
     """Test loading the Electricity dataset."""
@@ -324,3 +383,23 @@ def test_londonpassenger_loader():
         "distance",
     ]
     assert londonpassenger.shared_features_by_choice_names[0] == expected_shared_features_names
+
+
+def test_description():
+    """Test getting description."""
+    _ = load_swissmetro(return_desc=True)
+    _ = load_modecanada(return_desc=True)
+    _ = load_heating(return_desc=True)
+    _ = load_electricity(return_desc=True)
+    _ = load_train(return_desc=True)
+    _ = load_car_preferences(return_desc=True)
+    _ = load_hc(return_desc=True)
+    _ = load_londonpassenger(return_desc=True)
+    _ = load_tafeng(return_desc=True)
+
+
+def test_load_csv():
+    """Test csv file loader."""
+    _ = load_csv(data_file_name="test_data.csv", data_module="tests/data")
+    names, data = load_gzip("swissmetro.csv.gz", data_module="choice_learn/datasets/data")
+    _ = slice_from_names(data, names[:4], names)