rename conftest to helperfunctions

Author: niekdejonge

tests/conftest.py tests/helper_functions.pytests/conftest.py renamed to tests/helper_functions.py

@@ -3,7 +3,7 @@
 from matchms.filtering.metadata_processing.add_fingerprint import _derive_fingerprint_from_inchi
 from ms2query.benchmarking.AnnotatedSpectrumSet import AnnotatedSpectrumSet
 from ms2query.benchmarking.Fingerprints import Fingerprints, get_similarity_matrix
-from tests.conftest import create_test_spectra, get_inchikey_inchi_pairs
+from tests.helper_functions import create_test_spectra, get_inchikey_inchi_pairs
 
 
 @pytest.fixture

tests/test_benchmarking/test_Fingerprints.py

@@ -1,14 +1,15 @@
 from ms2query.benchmarking.AnnotatedSpectrumSet import (
     AnnotatedSpectrumSet,
 )
-from tests.conftest import create_test_spectra, ms2deepscore_model
+from tests.helper_functions import create_test_spectra, ms2deepscore_model
 
 
 def test_create_annotated_spectrum_set():
     test_spectra = create_test_spectra(nr_of_inchikeys=3, number_of_spectra_per_inchikey=3)
     spectrum_set = AnnotatedSpectrumSet.create_spectrum_set(spectra=test_spectra)
     assert len(spectrum_set.spectrum_indices_per_inchikey) == 3
 
+
 def test_add_spectrum_sets():
     test_spectra = create_test_spectra(nr_of_inchikeys=3, number_of_spectra_per_inchikey=3)
 
@@ -26,6 +27,7 @@ def test_add_spectrum_sets():
     combined_spectra = spectrum_set_1 + spectrum_set_2
     assert correct_combined_set == combined_spectra
 
+
 def test_subsetting():
     test_spectra = create_test_spectra(nr_of_inchikeys=3, number_of_spectra_per_inchikey=3)
 
@@ -37,4 +39,4 @@ def test_subsetting():
     correct_subsetted_set.add_embeddings(model)
     spectrum_set.add_embeddings(model)
 
-    assert correct_subsetted_set == spectrum_set.subset_spectra([0,1,2,3,4])
+    assert correct_subsetted_set == spectrum_set.subset_spectra([0, 1, 2, 3, 4])

tests/test_benchmarking/test_SpectrumDataSet.py

@@ -1,6 +1,6 @@
 import pytest
 from ms2query.benchmarking.Embeddings import Embeddings
-from tests.conftest import create_test_spectra, ms2deepscore_model
+from tests.helper_functions import create_test_spectra, ms2deepscore_model
 
 
 def test_subset_embeddings():
@@ -29,5 +29,3 @@ def test_combine_embeddings():
     assert combined_embeddings == correct_combined_embeddings
     with pytest.raises(ValueError):
         Embeddings.combine_embeddings(embeddings_1, embeddings_1)
-
-

tests/test_benchmarking/test_embeddings.py

@@ -1,6 +1,6 @@
 from ms2query.benchmarking.AnnotatedSpectrumSet import AnnotatedSpectrumSet
 from ms2query.benchmarking.EvaluateAnalogueSearch import EvaluateAnalogueSearch
-from tests.conftest import create_test_spectra, ms2deepscore_model
+from tests.helper_functions import create_test_spectra, ms2deepscore_model
 
 
 def create_dummy_library_and_validation_spectra() -> tuple[AnnotatedSpectrumSet, AnnotatedSpectrumSet]:

tests/test_benchmarking/test_evaluate_methods.py

@@ -10,7 +10,7 @@
     integrated_similarity_flow,
     predict_with_integrated_similarity_flow,
 )
-from tests.conftest import create_test_spectra, ms2deepscore_model
+from tests.helper_functions import create_test_spectra, ms2deepscore_model
 
 
 def get_library_and_test_spectra_not_identical() -> tuple[AnnotatedSpectrumSet, AnnotatedSpectrumSet]:

tests/test_benchmarking/test_methods.py

@@ -0,0 +1,30 @@
+import numpy as np
+import pytest
+from ms2query.benchmarking.AnnotatedSpectrumSet import AnnotatedSpectrumSet
+from ms2query.benchmarking.reference_methods.PredictMS2DeepScoreSimilarity import (
+    predict_top_ms2deepscores,
+)
+from tests.helper_functions import create_test_spectra, ms2deepscore_model
+
+
+@pytest.mark.parametrize(
+    "method",
+    [
+        predict_top_ms2deepscores,
+    ],
+)
+def test_predict_highest_ms2deepscore_similarity(method):
+    ms2ds_model = ms2deepscore_model()
+    test_spectra = create_test_spectra(1)
+    library_spectra = AnnotatedSpectrumSet.create_spectrum_set(test_spectra)
+    query_spectra = AnnotatedSpectrumSet.create_spectrum_set(test_spectra)
+    library_spectra.add_embeddings(ms2ds_model)
+    query_spectra.add_embeddings(ms2ds_model)
+    number_of_analogues = 2
+
+    indices, distances = method(library_spectra.embeddings, query_spectra.embeddings, k=number_of_analogues)
+    assert indices.shape == (len(test_spectra), number_of_analogues)
+    assert distances.shape == (len(test_spectra), number_of_analogues)
+    for i, row in enumerate(indices):
+        assert row[0] == i, "The highest predictions should be against itself"
+        assert np.allclose(distances[i][0], 1.0, atol=1e-5)