Merge pull request #813 from cmu-delphi/nowcast-nodata

Add better no data handling to nowcast

Author: krivard

nowcast/delphi_nowcast/data_containers.py

@@ -42,11 +42,15 @@ def add_data(self,
     @property
     def dates(self) -> list:
         """Check if there is no stored data in the class."""
+        if not self.data:
+            raise ValueError("No data")
         return list(self.data.keys())
 
     @property
     def values(self) -> list:
         """Check if there is no stored data in the class."""
+        if not self.data:
+            raise ValueError("No data")
         return list(self.data.values())
 
     def get_data_range(self,

nowcast/delphi_nowcast/sensorization/regression_model.py

@@ -39,17 +39,17 @@ def compute_regression_sensor(day: date,
         Float value of sensor on `date`
     """
     previous_day = day - timedelta(1)
-    first_day = max(min(covariate.dates), min(response.dates))
     try:
+        first_day = max(min(covariate.dates), min(response.dates))
         train_Y = response.get_data_range(first_day, previous_day)
         train_covariates = covariate.get_data_range(first_day, previous_day)
     except ValueError:
         return np.nan
     if not train_Y:
         return np.nan
-    train_Y, train_covariates = zip(  # only get pairs where both are not nan
-        *[(i, j) for i, j in zip(train_Y, train_covariates) if not (np.isnan(i) or np.isnan(j))]
-    )
+    non_nan_values = [(i, j) for i, j in zip(train_Y, train_covariates)
+                      if not (np.isnan(i) or np.isnan(j))]
+    train_Y, train_covariates = zip(*non_nan_values) if non_nan_values else ([], [])
     if len(train_Y) < MIN_SAMPLE_SIZE:
         print("insufficient observations")
         return np.nan

nowcast/tests/sensorization/test_regression_model.py

@@ -68,3 +68,8 @@ def test_compute_regression_sensor_out_of_range(self):
                   date(2020, 1, 7): 42}
         )
         assert np.isnan(compute_regression_sensor(date(2020, 1, 16), test_covariate, test_response, False))
+
+    def test_compute_regression_sensor_no_data(self):
+        test_covariate = LocationSeries()
+        test_response = LocationSeries()
+        assert np.isnan(compute_regression_sensor(date(2020, 1, 16), test_covariate, test_response, False))

nowcast/tests/test_data_containers.py

@@ -37,3 +37,10 @@ def test_get_data_range_invalid_impute(self):
         test_ls = LocationSeries(data={date(2020, 1, 1): 7, date(2020, 1, 2): np.nan, date(2020, 1, 3): 9})
         with pytest.raises(ValueError, match="Invalid imputation method. Must be None or 'mean'"):
             test_ls.get_data_range(date(2020, 1, 1), date(2020, 1, 3), "fakeimpute")
+
+    def test_no_data(self):
+        test_ls = LocationSeries()
+        with pytest.raises(ValueError, match="No data"):
+            test_ls.dates
+        with pytest.raises(ValueError, match="No data"):
+            test_ls.values