Organize pytest fixture

tests/conftest.py

@@ -0,0 +1,30 @@
+import json
+import pickle
+
+import pytest
+
+from clin_nlp_metrics import Dataset
+
+
+@pytest.fixture
+def mctrainer_data():
+    with open("tests/data/medcattrainer_export.json", "rb") as f:
+        return json.load(f)
+
+
+@pytest.fixture
+def mctrainer_dataset(mctrainer_data):
+    return Dataset.from_medcattrainer(data=mctrainer_data)
+
+
+@pytest.fixture
+def clinlp_docs():
+    with open("tests/data/clinlp_docs.pickle", "rb") as f:
+        return pickle.load(f)
+
+
+@pytest.fixture
+def clinlp_dataset(clinlp_docs):
+    ids = list(f"doc_{x}" for x in range(0, 15))
+
+    return Dataset.from_clinlp_docs(nlp_docs=clinlp_docs, ids=ids)

tests/test_dataset.py

@@ -1,29 +1,8 @@
-import json
-import pickle
-
 import clinlp  # noqa: F401
-import pytest
 
 from clin_nlp_metrics.dataset import Annotation, Dataset, Document
 
 
-@pytest.fixture
-def mctrainer_data():
-    with open("tests/data/medcattrainer_export.json", "rb") as f:
-        return json.load(f)
-
-
-@pytest.fixture
-def dataset(mctrainer_data):
-    return Dataset.from_medcattrainer(data=mctrainer_data)
-
-
-@pytest.fixture
-def clinlp_docs():
-    with open("tests/data/clinlp_docs.pickle", "rb") as f:
-        return pickle.load(f)
-
-
 class TestAnnotation:
     def test_annotation_nervaluate(self):
         ann = Annotation(text="test", start=0, end=5, label="test")
@@ -243,19 +222,19 @@ def test_dataset_nervaluate(self):
             [{"text": "test2", "start": 0, "end": 5, "label": "test2"}],
         ]
 
-    def test_dataset_to_nervaluate_with_filter(self, dataset):
+    def test_dataset_to_nervaluate_with_filter(self, mctrainer_dataset):
         def ann_filter(ann):
             return any(not qualifier["is_default"] for qualifier in ann.qualifiers)
 
-        to_nervaluate = dataset.to_nervaluate(ann_filter=ann_filter)
+        to_nervaluate = mctrainer_dataset.to_nervaluate(ann_filter=ann_filter)
 
         assert to_nervaluate[0] == [
             {"end": 23, "label": "C0002871_anemie", "start": 17, "text": "anemie"}
         ]
         assert to_nervaluate[1] == []
 
-    def test_infer_default_qualifiers(self, dataset):
-        default_qualifiers = dataset.infer_default_qualifiers()
+    def test_infer_default_qualifiers(self, mctrainer_dataset):
+        default_qualifiers = mctrainer_dataset.infer_default_qualifiers()
 
         assert default_qualifiers == {
             "Negation": "Affirmed",
@@ -264,71 +243,77 @@ def test_infer_default_qualifiers(self, dataset):
             "Plausibility": "Plausible",
         }
 
-    def test_num_docs(self, dataset):
-        assert dataset.num_docs() == 14
+    def test_num_docs(self, mctrainer_dataset):
+        assert mctrainer_dataset.num_docs() == 14
 
-    def test_num_annotations(self, dataset):
-        assert dataset.num_annotations() == 13
+    def test_num_annotations(self, mctrainer_dataset):
+        assert mctrainer_dataset.num_annotations() == 13
 
-    def test_span_counts(self, dataset):
-        assert len(dataset.span_counts()) == 11
+    def test_span_counts(self, mctrainer_dataset):
+        assert len(mctrainer_dataset.span_counts()) == 11
 
-    def test_span_counts_n_spans(self, dataset):
-        assert dataset.span_counts(n_spans=3) == {
+    def test_span_counts_n_spans(self, mctrainer_dataset):
+        assert mctrainer_dataset.span_counts(n_spans=3) == {
             "anemie": 2,
             "bloeding": 2,
             "prematuriteit": 1,
         }
 
-    def test_span_counts_callback(self, dataset):
-        assert dataset.span_counts(n_spans=3, span_callback=lambda x: x.upper()) == {
+    def test_span_counts_callback(self, mctrainer_dataset):
+        assert mctrainer_dataset.span_counts(
+            n_spans=3, span_callback=lambda x: x.upper()
+        ) == {
             "ANEMIE": 2,
             "BLOEDING": 2,
             "PREMATURITEIT": 1,
         }
 
-    def test_label_counts(self, dataset):
-        assert len(dataset.label_counts()) == 9
+    def test_label_counts(self, mctrainer_dataset):
+        assert len(mctrainer_dataset.label_counts()) == 9
 
-    def test_label_counts_n_labels(self, dataset):
-        assert dataset.label_counts(n_labels=3) == {
+    def test_label_counts_n_labels(self, mctrainer_dataset):
+        assert mctrainer_dataset.label_counts(n_labels=3) == {
             "C0002871_anemie": 2,
             "C0151526_prematuriteit": 2,
             "C0270191_intraventriculaire_bloeding": 2,
         }
 
-    def test_label_counts_callback(self, dataset):
-        assert dataset.label_counts(
+    def test_label_counts_callback(self, mctrainer_dataset):
+        assert mctrainer_dataset.label_counts(
             n_labels=3, label_callback=lambda x: x[x.index("_") + 1 :]
         ) == {"anemie": 2, "prematuriteit": 2, "intraventriculaire_bloeding": 2}
 
-    def test_qualifier_counts(self, dataset):
-        assert dataset.qualifier_counts() == {
+    def test_qualifier_counts(self, mctrainer_dataset):
+        assert mctrainer_dataset.qualifier_counts() == {
             "Experiencer": {"Patient": 12, "Other": 1},
             "Negation": {"Affirmed": 11, "Negated": 2},
             "Plausibility": {"Plausible": 11, "Hypothetical": 2},
             "Temporality": {"Current": 11, "Historical": 2},
         }
 
-    def test_stats(self, dataset):
-        stats = dataset.stats()
+    def test_stats(self, mctrainer_dataset):
+        stats = mctrainer_dataset.stats()
 
-        assert stats["num_docs"] == dataset.num_docs()
-        assert stats["num_annotations"] == dataset.num_annotations()
-        assert stats["span_counts"] == dataset.span_counts()
-        assert stats["label_counts"] == dataset.label_counts()
-        assert stats["qualifier_counts"] == dataset.qualifier_counts()
+        assert stats["num_docs"] == mctrainer_dataset.num_docs()
+        assert stats["num_annotations"] == mctrainer_dataset.num_annotations()
+        assert stats["span_counts"] == mctrainer_dataset.span_counts()
+        assert stats["label_counts"] == mctrainer_dataset.label_counts()
+        assert stats["qualifier_counts"] == mctrainer_dataset.qualifier_counts()
 
-    def test_stats_with_kwargs(self, dataset):
+    def test_stats_with_kwargs(self, mctrainer_dataset):
         n_labels = 1
         span_callback = lambda x: x.upper()  # noqa: E731
 
-        stats = dataset.stats(
+        stats = mctrainer_dataset.stats(
             n_labels=n_labels, span_callback=span_callback, unused_argument=None
         )
 
-        assert stats["num_docs"] == dataset.num_docs()
-        assert stats["num_annotations"] == dataset.num_annotations()
-        assert stats["span_counts"] == dataset.span_counts(span_callback=span_callback)
-        assert stats["label_counts"] == dataset.label_counts(n_labels=n_labels)
-        assert stats["qualifier_counts"] == dataset.qualifier_counts()
+        assert stats["num_docs"] == mctrainer_dataset.num_docs()
+        assert stats["num_annotations"] == mctrainer_dataset.num_annotations()
+        assert stats["span_counts"] == mctrainer_dataset.span_counts(
+            span_callback=span_callback
+        )
+        assert stats["label_counts"] == mctrainer_dataset.label_counts(
+            n_labels=n_labels
+        )
+        assert stats["qualifier_counts"] == mctrainer_dataset.qualifier_counts()

tests/test_metrics.py

@@ -1,27 +1,6 @@
-import json
-import pickle
-
 import pytest
 
-from clin_nlp_metrics import Dataset, Metrics
-
-
-@pytest.fixture
-def mctrainer_dataset():
-    with open("tests/data/medcattrainer_export.json", "rb") as f:
-        mctrainer_data = json.load(f)
-
-    return Dataset.from_medcattrainer(data=mctrainer_data)
-
-
-@pytest.fixture
-def clinlp_dataset():
-    with open("tests/data/clinlp_docs.pickle", "rb") as f:
-        data = pickle.load(f)
-
-    ids = list(f"doc_{x}" for x in range(0, 15))
-
-    return Dataset.from_clinlp_docs(nlp_docs=data, ids=ids)
+from clin_nlp_metrics import Metrics
 
 
 class TestMetrics: