Merge pull request #1860 from cmu-delphi/release/indicators_v0.3.41_utils_v0.3.16

Release covidcast-indicators 0.3.41

Author: korlaxxalrok

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.3.40
+current_version = 0.3.41
 commit = True
 message = chore: bump covidcast-indicators to {new_version}
 tag = False

changehc/delphi_changehc/run.py

@@ -78,6 +78,16 @@ def make_asserts(params):
         assert (files["denom"] is None) == (files["flu"] is None), \
             "exactly one of denom and flu files are provided"
 
+def process_dates(params, startdate_dt, enddate_dt):
+    """Process the start and end dates for indicator."""
+    enddate = params["indicator"].get("end_date")
+    if enddate is None:
+        enddate = str(enddate_dt.date())
+    startdate = params["indicator"].get("start_date", str(startdate_dt.date()))
+    if startdate is None:
+        startdate = str(startdate_dt.date())
+    return startdate, enddate
+
 
 def run_module(params: Dict[str, Dict[str, Any]]):
     """
@@ -144,8 +154,8 @@ def run_module(params: Dict[str, Dict[str, Any]]):
     enddate_dt = dropdate_dt - timedelta(days=n_waiting_days)
     startdate_dt = enddate_dt - timedelta(days=n_backfill_days)
     # now allow manual overrides
-    enddate = enddate = params["indicator"].get("end_date",str(enddate_dt.date()))
-    startdate = params["indicator"].get("start_date", str(startdate_dt.date()))
+
+    startdate, enddate = process_dates(params, startdate_dt, enddate_dt)
 
     logger.info("generating signal and exporting to CSV",
         first_sensor_date = startdate,

changehc/delphi_changehc/update_sensor.py

@@ -160,6 +160,8 @@ def geo_reindex(self, data):
                                                  thr_col="den",
                                                  mega_col=geo,
                                                  date_col=Config.DATE_COL)
+            # this line should be removed once the fix is implemented for megacounties
+            data_frame = data_frame[~((data_frame['county'].str.len() > 5) | (data_frame['county'].str.contains('_')))]
         elif geo == "state":
             data_frame = gmpr.replace_geocode(data, "fips", "state_id", new_col="state",
                                               date_col=Config.DATE_COL)
@@ -172,7 +174,8 @@ def geo_reindex(self, data):
         multiindex = pd.MultiIndex.from_product((unique_geo_ids, self.fit_dates),
                                                 names=[geo, Config.DATE_COL])
         assert (len(multiindex) <= (len(gmpr.get_geo_values(gmpr.as_mapper_name(geo))) * len(self.fit_dates))
-                ), "more loc-date pairs than maximum number of geographies x number of dates"
+                ), f"more loc-date pairs than maximum number of geographies x number of dates, length of multiindex is {len(multiindex)}, geo level is {geo}"
+
         # fill dataframe with missing dates using 0
         data_frame = data_frame.reindex(multiindex, fill_value=0)
         data_frame.fillna(0, inplace=True)

changehc/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41

claims_hosp/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41

doctor_visits/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41

dsew_community_profile/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41

google_symptoms/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41

hhs_hosp/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41

nchs_mortality/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41

nowcast/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41

quidel_covidtest/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41

sir_complainsalot/version.cfg

@@ -1 +1 @@
-current_version = 0.3.40
+current_version = 0.3.41