test_metric with pytest.mark.parametrize

CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog of galpynostatic
 
+## v0.2.2 (2023-12-27)
+
+### Bug fixes
+
+- Change the name of the `bmx_fc` metric to `umbem`, which is the name of the metric in the cited PhD thesis.
+- Replace the `test_metric` to use `pytest.mark.parametrize` and have a test for each value insted of the dataframe all togheter.
+
 ## v0.2.1 (2023-12-26)
 
 ### Bug fixes

tests/test_metric.py

@@ -17,40 +17,72 @@
 
 import numpy as np
 
-import pandas as pd
-
 import pytest
 
 # =============================================================================
 # TESTS
 # =============================================================================
 
 
-def test_fom(data_path):
-    """Test the Figure of Merit for fast-charging of Xia et al."""
-    df = pd.read_csv(data_path / "metric_dataset.csv")
-
-    for d, dcoeff, ref in zip(df["d"], df["dcoeff"], df["ref_tau"]):
+@pytest.mark.parametrize(
+    ("d", "dcoeff", "ref"),
+    [
+        (10.0, 7.8e-11, [0.101967, 12820.512821]),
+        (8.0, 3.324154e-10, [0.599743, 1925.301896]),
+        (3.5, 1.732051e-11, [0.258640, 7072.540012]),
+        (7.5, 1.0e-09, [0.778005, 562.5]),
+        (2.775, 3.872983e-10, [0.917220, 198.8293]),
+        (0.5, 1.0e-09, [0.986813, 2.5]),
+        (3.0, 8.246211e-11, [0.802635, 1091.410346]),
+        (0.1, 1.34e-11, [0.99418, 7.462687]),
+        (0.2, 9.04e-11, [0.992554, 4.424779]),
+        (3.5, 1.5e-12, [0.022498, 81666.666667]),
+        (0.3, 2.97e-12, [0.953725, 303.030303]),
+        (3.5, 1.57e-07, [0.927188, 0.780255]),
+        (10.0, 1.0e-08, [0.785281, 100.0]),
+        (0.1, 1.07e-12, [0.983569, 93.457944]),
+        (0.2, 1.486607e-14, [0.076591, 26906.909493]),
+        (0.35, 8.726168e-12, [0.972794, 140.382353]),
+        (0.3, 9.872386e-15, [0.023048, 91163.372259]),
+        (0.2, 3.640055e-14, [0.184536, 10988.84494877138]),
+        (0.3, 4.547417e-12, [0.966699, 197.914552]),
+        (0.45, 4.0e-12, [0.925663, 506.25]),
+        (0.935, 4.690416e-13, [0.106935, 18638.538671]),
+        (0.2, 1.03e-13, [0.525938, 3883.495146]),
+        (0.09, 1.04e-13, [0.899222, 778.846154]),
+        (0.935, 1.0e-11, [0.870576, 874.225000]),
+        (15.0, 1.430315e-10, [0.064599, 15730.800558]),
+        (10.0, 4.690416e-10, [0.534568, 2132.007054]),
+        (25.166667, 7.6e-11, [0.006433, 83336.990512]),
+        (25.166667, 4.4e-06, [0.493498, 1.439457]),
+        (50.5, 1.0e-07, [0.0, 255.025]),
+    ],
+)
+class TestMetric:
+    """Test the different metrics for benchmarking FC materials."""
+
+    def test_fom(self, d, dcoeff, ref):
+        """Test the Figure of Merit for fast-charging of Xia et al."""
         value = galpynostatic.metric.fom(1e-4 * d, dcoeff)
 
-        np.testing.assert_almost_equal(value, ref, 6)
+        np.testing.assert_almost_equal(value, ref[1], 6)
 
+    def test_umbem(self, d, dcoeff, ref):
+        """Test the UMBEM metric without fitting the model."""
+        greg = galpynostatic.model.GalvanostaticRegressor(d=1e-4 * d)
+        greg._validate_geometry()
+        greg._map = galpynostatic.datasets.map.MapSpline(greg.dataset)
+        greg.dcoeff_, greg.k0_ = dcoeff, 1e-7
+        greg.dcoeff_err_ = None
+
+        value = galpynostatic.metric.umbem(greg)
 
-@pytest.mark.parametrize(("fit"), [(True), (False)])
-def test_umbem(fit, data_path):
-    """Test the Benchmarking Materials for eXtreme fast-charging metric."""
-    df = pd.read_csv(data_path / "metric_dataset.csv")
+        np.testing.assert_almost_equal(value, ref[0], 6)
 
-    for d, dcoeff, ref in zip(df["d"], df["dcoeff"], df["ref_soc"]):
-        if fit:
-            greg = {"d": 1e-4 * d, "dcoeff_": dcoeff, "k0_": 1e-7}
-        else:
-            greg = galpynostatic.model.GalvanostaticRegressor(d=1e-4 * d)
-            greg._validate_geometry()
-            greg._map = galpynostatic.datasets.map.MapSpline(greg.dataset)
-            greg.dcoeff_, greg.k0_ = dcoeff, 1e-7
-            greg.dcoeff_err_ = None
+    def test_umbem_fit(self, d, dcoeff, ref):
+        """Test the UMBEM metric using the fit argument."""
+        greg = {"d": 1e-4 * d, "dcoeff_": dcoeff, "k0_": 1e-7}
 
         value = galpynostatic.metric.umbem(greg)
 
-        np.testing.assert_almost_equal(value, ref, 6)
+        np.testing.assert_almost_equal(value, ref[0], 6)