Test furness/no furness in jacobian

Transport-for-the-North · Dec 12, 2023 · 3278725 · 3278725
1 parent d3b3cb3
commit 3278725
Showing 1 changed file with 45 additions and 6 deletions.
diff --git a/tests/gravity_model/test_multi_area.py b/tests/gravity_model/test_multi_area.py
@@ -77,7 +77,7 @@ def fixture_dists(data_dir, distributions):
     )
     return sorted
 
-@pytest.fixture(name="multi_conf", scope="session")
+@pytest.fixture(name="no_furness_jac_conf", scope="session")
 def fixture_conf(data_dir, mock_dir):
     conf = gm.MultiDistInput(TLDFile=data_dir / 'distributions.csv',
                              TldLookupFile=data_dir / 'distributions_lookup.csv',
@@ -91,20 +91,55 @@ def fixture_conf(data_dir, mock_dir):
                              init_params={'mu': 1, 'sigma': 2},
                              log_path=mock_dir / 'log.csv',
                              furness_tolerance=0.1,
+                             furness_jac=False
                              )
     return conf
 
+@pytest.fixture(name="furness_jac_conf", scope="session")
+def fixture_jac_furn(data_dir, mock_dir):
+    conf = gm.MultiDistInput(TLDFile=data_dir / 'distributions.csv',
+                             TldLookupFile=data_dir / 'distributions_lookup.csv',
+                             cat_col='cat',
+                             min_col='lower',
+                             max_col='upper',
+                             ave_col='avg',
+                             trips_col='trips',
+                             lookup_cat_col='cat',
+                             lookup_zone_col='zone',
+                             init_params={'mu': 1, 'sigma': 2},
+                             log_path=mock_dir / 'log.csv',
+                             furness_tolerance=0.1,
+                             furness_jac=True,
+                             )
+    return conf
+
+
+@pytest.fixture(name="cal_no_furness", scope="session")
+def fixture_cal_no_furness(data_dir, infilled, no_furness_jac_conf, trip_ends, mock_dir):
+    row_targets = trip_ends["origin"].values
+    col_targets = trip_ends["destination"].values
+    model = gm.MultiAreaGravityModelCalibrator(
+        row_targets=row_targets,
+        col_targets=col_targets,
+        cost_matrix=infilled,
+        cost_function=cost_functions.BuiltInCostFunction.LOG_NORMAL.get_cost_function(),
+        params=no_furness_jac_conf
+    )
+    results = model.calibrate(
+        running_log_path=mock_dir / "temp_log.csv"
+    )
+    return results
 
-@pytest.fixture(name="cal_results", scope="session")
-def fixture_cal_res(data_dir, infilled, multi_conf, trip_ends, mock_dir):
+@pytest.fixture(name="cal_furness", scope="session")
+def fixture_cal_furness(data_dir, infilled, furness_jac_conf, trip_ends, mock_dir):
     row_targets = trip_ends["origin"].values
     col_targets = trip_ends["destination"].values
     model = gm.MultiAreaGravityModelCalibrator(
         row_targets=row_targets,
         col_targets=col_targets,
         cost_matrix=infilled,
         cost_function=cost_functions.BuiltInCostFunction.LOG_NORMAL.get_cost_function(),
-        params=multi_conf
+        params=furness_jac_conf
     )
     results = model.calibrate(
         running_log_path=mock_dir / "temp_log.csv"
@@ -125,12 +160,16 @@ def test_tld_processing_type(self, distributions_sorted):
 
 class TestDist:
     @pytest.mark.parametrize("area", ['City', 'Town', 'External', 'Village'])
-    def test_convergence(self, cal_results, area):
+    @pytest.mark.parametrize("cal_results", ['cal_furness', 'cal_no_furness'])
+    def test_convergence(self, cal_results, area, request):
+        cal_results = request.getfixturevalue(cal_results)
         dist = cal_results[area]
         assert dist.cost_convergence > 0.85
 
     @pytest.mark.parametrize("area", ['City', 'Town', 'External', 'Village'])
-    def test_params(self, cal_results, area):
+    @pytest.mark.parametrize("cal_results", ['cal_furness', 'cal_no_furness'])
+    def test_params(self, cal_results, area, request):
+        cal_results=request.getfixturevalue(cal_results)
         dist = cal_results[area]
         mu = dist.cost_params['mu']
         sigma = dist.cost_params['sigma']