test: integrate data into test files

Safe-DS · Mar 29, 2023 · 9e7b7fe · 9e7b7fe
1 parent c2edf88
commit 9e7b7fe
Show file tree

Hide file tree

Showing 35 changed files with 249 additions and 151 deletions.
diff --git a/tests/resources/test_decision_tree.csv b/tests/resources/test_decision_tree.csv
diff --git a/tests/resources/test_decision_tree_invalid.csv b/tests/resources/test_decision_tree_invalid.csv
diff --git a/tests/resources/test_elastic_net_regression.csv b/tests/resources/test_elastic_net_regression.csv
diff --git a/tests/resources/test_elastic_net_regression_invalid.csv b/tests/resources/test_elastic_net_regression_invalid.csv
diff --git a/tests/resources/test_gradient_boosting_classification.csv b/tests/resources/test_gradient_boosting_classification.csv
diff --git a/tests/resources/test_gradient_boosting_classification_invalid.csv b/tests/resources/test_gradient_boosting_classification_invalid.csv
diff --git a/tests/resources/test_gradient_boosting_regression.csv b/tests/resources/test_gradient_boosting_regression.csv
diff --git a/tests/resources/test_gradient_boosting_regression_invalid.csv b/tests/resources/test_gradient_boosting_regression_invalid.csv
diff --git a/tests/resources/test_k_nearest_neighbors.csv b/tests/resources/test_k_nearest_neighbors.csv
diff --git a/tests/resources/test_k_nearest_neighbors_invalid.csv b/tests/resources/test_k_nearest_neighbors_invalid.csv
diff --git a/tests/resources/test_lasso_regression.csv b/tests/resources/test_lasso_regression.csv
diff --git a/tests/resources/test_lasso_regression_invalid.csv b/tests/resources/test_lasso_regression_invalid.csv
diff --git a/tests/resources/test_linear_regression.csv b/tests/resources/test_linear_regression.csv
diff --git a/tests/resources/test_linear_regression_invalid.csv b/tests/resources/test_linear_regression_invalid.csv
diff --git a/tests/resources/test_logistic_regression.csv b/tests/resources/test_logistic_regression.csv
diff --git a/tests/resources/test_logistic_regression_invalid.csv b/tests/resources/test_logistic_regression_invalid.csv
diff --git a/tests/resources/test_random_forest.csv b/tests/resources/test_random_forest.csv
diff --git a/tests/resources/test_random_forest_invalid.csv b/tests/resources/test_random_forest_invalid.csv
diff --git a/tests/resources/test_ridge_regression.csv b/tests/resources/test_ridge_regression.csv
diff --git a/tests/resources/test_ridge_regression_invalid.csv b/tests/resources/test_ridge_regression_invalid.csv
diff --git a/tests/safeds/ml/classification/test_classifier.py b/tests/safeds/ml/classification/test_classifier.py
@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import pandas as pd
+
 from safeds.data.tabular.containers import Column, Table, TaggedTable
 from safeds.ml.classification import Classifier
 

diff --git a/tests/safeds/ml/classification/test_decision_tree.py b/tests/safeds/ml/classification/test_decision_tree.py
@@ -1,8 +1,8 @@
 import pytest
-from safeds.data.tabular.containers import Table, TaggedTable
+
+from safeds.data.tabular.containers import Table, TaggedTable, Column
 from safeds.exceptions import LearningError, PredictionError
 from safeds.ml.classification import Classifier, DecisionTree
-from tests.helpers import resolve_resource_path
 
 
 @pytest.fixture()
@@ -12,14 +12,26 @@ def classifier() -> Classifier:
 
 @pytest.fixture()
 def valid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_decision_tree.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", [2, 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 @pytest.fixture()
 def invalid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_decision_tree_invalid.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", ["a", 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 class TestFit:
@@ -41,7 +53,7 @@ def test_should_include_features_of_prediction_input(self, classifier: Classifie
     def test_should_set_correct_target_name(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         fitted_classifier = classifier.fit(valid_data)
         prediction = fitted_classifier.predict(valid_data.features)
-        assert prediction.target.name == "T"
+        assert prediction.target.name == "target"
 
     def test_should_raise_when_not_fitted(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         with pytest.raises(PredictionError):

diff --git a/tests/safeds/ml/classification/test_gradient_boosting.py b/tests/safeds/ml/classification/test_gradient_boosting.py
@@ -1,8 +1,8 @@
 import pytest
-from safeds.data.tabular.containers import Table, TaggedTable
+
+from safeds.data.tabular.containers import Table, TaggedTable, Column
 from safeds.exceptions import LearningError, PredictionError
 from safeds.ml.classification import Classifier, GradientBoosting
-from tests.helpers import resolve_resource_path
 
 
 @pytest.fixture()
@@ -12,14 +12,26 @@ def classifier() -> Classifier:
 
 @pytest.fixture()
 def valid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_gradient_boosting_classification.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", [2, 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 @pytest.fixture()
 def invalid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_gradient_boosting_classification_invalid.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", ["a", 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 class TestFit:
@@ -41,7 +53,7 @@ def test_should_include_features_of_prediction_input(self, classifier: Classifie
     def test_should_set_correct_target_name(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         fitted_classifier = classifier.fit(valid_data)
         prediction = fitted_classifier.predict(valid_data.features)
-        assert prediction.target.name == "T"
+        assert prediction.target.name == "target"
 
     def test_should_raise_when_not_fitted(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         with pytest.raises(PredictionError):

diff --git a/tests/safeds/ml/classification/test_k_nearest_neighbors.py b/tests/safeds/ml/classification/test_k_nearest_neighbors.py
@@ -1,5 +1,6 @@
 import pytest
-from safeds.data.tabular.containers import Table, TaggedTable
+
+from safeds.data.tabular.containers import Table, TaggedTable, Column
 from safeds.exceptions import LearningError, PredictionError
 from safeds.ml.classification import Classifier, KNearestNeighbors
 from tests.helpers import resolve_resource_path
@@ -12,14 +13,26 @@ def classifier() -> Classifier:
 
 @pytest.fixture()
 def valid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_k_nearest_neighbors.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", [2, 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 @pytest.fixture()
 def invalid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_k_nearest_neighbors_invalid.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", ["a", 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 class TestFit:
@@ -41,7 +54,7 @@ def test_should_include_features_of_prediction_input(self, classifier: Classifie
     def test_should_set_correct_target_name(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         fitted_classifier = classifier.fit(valid_data)
         prediction = fitted_classifier.predict(valid_data.features)
-        assert prediction.target.name == "T"
+        assert prediction.target.name == "target"
 
     def test_should_raise_when_not_fitted(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         with pytest.raises(PredictionError):

diff --git a/tests/safeds/ml/classification/test_logistic_regression.py b/tests/safeds/ml/classification/test_logistic_regression.py
@@ -1,8 +1,8 @@
 import pytest
-from safeds.data.tabular.containers import Table, TaggedTable
+
+from safeds.data.tabular.containers import Table, TaggedTable, Column
 from safeds.exceptions import LearningError, PredictionError
 from safeds.ml.classification import Classifier, LogisticRegression
-from tests.helpers import resolve_resource_path
 
 
 @pytest.fixture()
@@ -12,14 +12,26 @@ def classifier() -> Classifier:
 
 @pytest.fixture()
 def valid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_logistic_regression.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", [2, 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 @pytest.fixture()
 def invalid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_logistic_regression_invalid.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", ["a", 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 class TestFit:
@@ -41,7 +53,7 @@ def test_should_include_features_of_prediction_input(self, classifier: Classifie
     def test_should_set_correct_target_name(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         fitted_classifier = classifier.fit(valid_data)
         prediction = fitted_classifier.predict(valid_data.features)
-        assert prediction.target.name == "T"
+        assert prediction.target.name == "target"
 
     def test_should_raise_when_not_fitted(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         with pytest.raises(PredictionError):

diff --git a/tests/safeds/ml/classification/test_random_forest.py b/tests/safeds/ml/classification/test_random_forest.py
@@ -1,8 +1,8 @@
 import pytest
-from safeds.data.tabular.containers import Table, TaggedTable
+
+from safeds.data.tabular.containers import Table, TaggedTable, Column
 from safeds.exceptions import LearningError, PredictionError
 from safeds.ml.classification import Classifier, RandomForest
-from tests.helpers import resolve_resource_path
 
 
 @pytest.fixture()
@@ -12,14 +12,26 @@ def classifier() -> Classifier:
 
 @pytest.fixture()
 def valid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_random_forest.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", [2, 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 @pytest.fixture()
 def invalid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_random_forest_invalid.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", ["a", 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 class TestFit:
@@ -41,7 +53,7 @@ def test_should_include_features_of_prediction_input(self, classifier: Classifie
     def test_should_set_correct_target_name(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         fitted_classifier = classifier.fit(valid_data)
         prediction = fitted_classifier.predict(valid_data.features)
-        assert prediction.target.name == "T"
+        assert prediction.target.name == "target"
 
     def test_should_raise_when_not_fitted(self, classifier: Classifier, valid_data: TaggedTable) -> None:
         with pytest.raises(PredictionError):

diff --git a/tests/safeds/ml/regression/test_decision_tree.py b/tests/safeds/ml/regression/test_decision_tree.py
@@ -1,8 +1,8 @@
 import pytest
-from safeds.data.tabular.containers import Table, TaggedTable
+
+from safeds.data.tabular.containers import Table, TaggedTable, Column
 from safeds.exceptions import LearningError, PredictionError
 from safeds.ml.regression import DecisionTree, Regressor
-from tests.helpers import resolve_resource_path
 
 
 @pytest.fixture()
@@ -12,14 +12,26 @@ def regressor() -> Regressor:
 
 @pytest.fixture()
 def valid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_decision_tree.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", [2, 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 @pytest.fixture()
 def invalid_data() -> TaggedTable:
-    table = Table.from_csv_file(resolve_resource_path("test_decision_tree_invalid.csv"))
-    return table.tag_columns(target_name="T")
+    return Table.from_columns(
+        [
+            Column("id", [1, 4]),
+            Column("feat1", ["a", 5]),
+            Column("feat2", [3, 6]),
+            Column("target", [0, 1]),
+        ]
+    ).tag_columns(target_name="target", feature_names=["feat1", "feat2"])
 
 
 class TestFit:
@@ -41,7 +53,7 @@ def test_should_include_features_of_prediction_input(self, regressor: Regressor,
     def test_should_set_correct_target_name(self, regressor: Regressor, valid_data: TaggedTable) -> None:
         fitted_regressor = regressor.fit(valid_data)
         prediction = fitted_regressor.predict(valid_data.features)
-        assert prediction.target.name == "T"
+        assert prediction.target.name == "target"
 
     def test_should_raise_when_not_fitted(self, regressor: Regressor, valid_data: TaggedTable) -> None:
         with pytest.raises(PredictionError):