diff --git a/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.ipynb b/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.ipynb
index f023b748..9d262e79 100644
--- a/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.ipynb
+++ b/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.ipynb
@@ -1084,12 +1084,14 @@
     "#    all available models are included by default)\n",
     "# - \"fold\" defines the number of folds to use for cross-validation.\n",
     "\n",
-    "# Note: This is only relevant if we are executing automated tests\n",
+    "# On CI/testing, only evaluate a single cheap model.\n",
+    "# # Alternatives: arima, ets, et_cds_dt, exp_smooth, naive.\n",
     "if \"PYTEST_CURRENT_TEST\" in os.environ:\n",
     "    best_models = compare_models(sort=\"MASE\",\n",
-    "                                 include=[\"ets\", \"et_cds_dt\", \"naive\"],\n",
-    "                                 n_select=3)\n",
-    "# If we are not in an automated test, compare all available models\n",
+    "                                 include=[\"ets\"],\n",
+    "                                 n_select=1)\n",
+    "\n",
+    "# When not on CI/testing, compare all available models.\n",
     "else:\n",
     "    best_models = compare_models(sort=\"MASE\", n_select=3)"
    ]
diff --git a/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.py b/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.py
index 06aa7b33..4d15f3ed 100644
--- a/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.py
+++ b/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.py
@@ -82,10 +82,15 @@ def fetch_data():
 
 def run_experiment(data):
     setup(data = data, fh=15, target="total_sales", index="month", log_experiment=True)
+
+    # On CI/testing, only evaluate a single cheap model.
+    # Alternatives: arima, ets, et_cds_dt, exp_smooth, naive.
     if "PYTEST_CURRENT_TEST" in os.environ:
         best_models = compare_models(sort="MASE",
-                                    include=["arima", "ets", "exp_smooth"],
-                                    n_select=3)
+                                     include=["ets"],
+                                     n_select=1)
+
+    # When not on CI/testing, compare all available models.
     else:
         best_models = compare_models(sort="MASE", n_select=3)