Reducing testing time (#1625)

* Marking slow tests

* Group torch tests

---------

Co-authored-by: Blazej Nowicki <blazej.nowicki@unit8.co>

darts/tests/datasets/test_dataset_loaders.py

@@ -1,5 +1,7 @@
 import os
 
+import pytest
+
 from darts import TimeSeries
 from darts.datasets import (
     _DEFAULT_PATH,
@@ -128,6 +130,7 @@ def tearDown(self):
             os.remove(os.path.join(default_directory, f))
         os.rmdir(DatasetLoader._DEFAULT_DIRECTORY)
 
+    @pytest.mark.slow
     def test_ok_dataset(self):
         for width, dataset_cls in zip(width_datasets, datasets):
             dataset = dataset_cls()

darts/tests/models/forecasting/test_TCN.py

@@ -1,3 +1,5 @@
+import pytest
+
 from darts.logging import get_logger
 from darts.metrics import mae
 from darts.tests.base_test_class import DartsBaseTestClass
@@ -65,6 +67,7 @@ def test_performance(self):
 
             self.assertTrue(mae(pred, test) < 0.3)
 
+        @pytest.mark.slow
         def test_coverage(self):
             torch.manual_seed(0)
             input_chunk_lengths = range(20, 50)

darts/tests/models/forecasting/test_historical_forecasts.py

@@ -471,6 +471,7 @@ def test_regression_auto_start_multiple_no_cov(self):
                     "of retrain=True and overlap_end=False, and a time index of type DateTimeIndex.",
                 )
 
+        @pytest.mark.slow
         @unittest.skipUnless(
             TORCH_AVAILABLE,
             "Torch not available. auto start and multiple time series for torch models will be skipped.",
@@ -637,6 +638,7 @@ def test_regression_auto_start_multiple_with_cov(self):
                     " retrain=True and overlap_end=False",
                 )
 
+        @pytest.mark.slow
         @unittest.skipUnless(
             TORCH_AVAILABLE,
             "Torch not available. auto start and multiple time series for torch models and covariates "

darts/tests/models/forecasting/test_local_forecasting_models.py

@@ -359,6 +359,7 @@ def test_forecast_time_index(self):
         self.assertEqual(pred.start_time(), pd.Timestamp("20130121"))
         self.assertEqual(pred.end_time(), pd.Timestamp("20130125"))
 
+    @pytest.mark.slow
     def test_statsmodels_future_models(self):
 
         # same tests, but VARIMA requires to work on a multivariate target series

darts/tests/models/forecasting/test_probabilistic_models.py

@@ -1,4 +1,5 @@
 import numpy as np
+import pytest
 
 from darts import TimeSeries
 from darts.logging import get_logger
@@ -135,6 +136,7 @@
     ]
 
 
+@pytest.mark.slow
 class ProbabilisticTorchModelsTestCase(DartsBaseTestClass):
     np.random.seed(0)
 
@@ -144,6 +146,7 @@ class ProbabilisticTorchModelsTestCase(DartsBaseTestClass):
     constant_noisy_multivar_ts = constant_noisy_ts.stack(constant_noisy_ts)
     num_samples = 5
 
+    @pytest.mark.slow
     def test_fit_predict_determinism(self):
 
         for model_cls, model_kwargs, _ in models_cls_kwargs_errs:

darts/tests/models/forecasting/test_regression_models.py

@@ -6,6 +6,7 @@
 
 import numpy as np
 import pandas as pd
+import pytest
 from sklearn.ensemble import HistGradientBoostingRegressor, RandomForestRegressor
 from sklearn.linear_model import LinearRegression
 from sklearn.preprocessing import OneHotEncoder
@@ -1191,6 +1192,7 @@ def test_models_runnability(self):
                 prediction = model_instance.predict(n=1)
                 self.assertEqual(len(prediction), 1)
 
+    @pytest.mark.slow
     def test_fit(self):
         multi_models_modes = [True, False]
         for mode in multi_models_modes:
@@ -2217,6 +2219,7 @@ class ProbabilisticRegressionModelsTestCase(DartsBaseTestClass):
     constant_noisy_multivar_ts = constant_noisy_ts.stack(constant_noisy_ts)
     num_samples = 5
 
+    @pytest.mark.slow
     def test_fit_predict_determinism(self):
         multi_models_modes = [False, True]
         for mode in multi_models_modes:
@@ -2237,6 +2240,7 @@ def test_fit_predict_determinism(self):
                 pred3 = model.predict(n=10, num_samples=2).values()
                 self.assertTrue((pred2 != pred3).any())
 
+    @pytest.mark.slow
     def test_probabilistic_forecast_accuracy(self):
         multi_models_modes = [True, False]
         for mode in multi_models_modes:

pytest.ini

@@ -0,0 +1,4 @@
+[pytest]
+addopts = --strict-markers
+markers =
+    slow: marks tests as slow (deselect with '-m "not slow"')