Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
32 changes: 19 additions & 13 deletions claims_hosp/delphi_claims_hosp/backfill.py
Original file line number Diff line number Diff line change
Expand Up @@ -36,23 +36,29 @@ def store_backfill_file(claims_filepath, _end_date, backfill_dir):
dtype=Config.CLAIMS_DTYPES,
parse_dates=[Config.CLAIMS_DATE_COL],
)
backfilldata.rename({"ServiceDate": "time_value",
"PatCountyFIPS": "fips",
"Denominator": "den",
"Covid_like": "num"},
axis=1, inplace=True)
backfilldata = gmpr.add_geocode(backfilldata, from_code="fips", new_code="state_id",
from_col="fips", new_col="state_id")
backfilldata.rename(
{
"ServiceDate": "time_value",
"PatCountyFIPS": "fips",
"Denominator": "den",
"Covid_like": "num",
"Flu1": "num_flu",
},
axis=1,
inplace=True,
)
backfilldata = gmpr.add_geocode(
backfilldata, from_code="fips", new_code="state_id", from_col="fips", new_col="state_id"
)
#Store one year's backfill data
if _end_date.day == 29 and _end_date.month == 2:
_start_date = datetime(_end_date.year-1, 2, 28)
else:
_start_date = _end_date.replace(year=_end_date.year-1)
selected_columns = ['time_value', 'fips', 'state_id',
'den', 'num']
backfilldata = backfilldata.loc[(backfilldata["time_value"] >= _start_date)
& (~backfilldata["fips"].isnull()),
selected_columns]
_start_date = _end_date.replace(year=_end_date.year - 1)
selected_columns = ["time_value", "fips", "state_id", "den", "num", "num_flu"]
backfilldata = backfilldata.loc[
(backfilldata["time_value"] >= _start_date) & (~backfilldata["fips"].isnull()), selected_columns
]

backfilldata["lag"] = [(_end_date - x).days for x in backfilldata["time_value"]]
backfilldata["time_value"] = backfilldata.time_value.dt.strftime("%Y-%m-%d")
Expand Down
13 changes: 10 additions & 3 deletions claims_hosp/delphi_claims_hosp/config.py
Original file line number Diff line number Diff line change
Expand Up @@ -13,8 +13,14 @@
class Config:
"""Static configuration variables."""

signal_name = "smoothed_covid19_from_claims"
signal_weekday_name = "smoothed_adj_covid19_from_claims"
signal_name = {
"Covid_like": "smoothed_covid19_from_claims",
"Flu1": "smoothed_flu_from_claims",
}
signal_weekday_name = {
"Covid_like": "smoothed_adj_covid19_from_claims",
"Flu1": "smoothed_adj_flu_from_claims",
}

# max number of CPUs available for pool
MAX_CPU_POOL = 10
Expand All @@ -30,7 +36,7 @@ class Config:
DAY_SHIFT = timedelta(days=0)

# data columns
CLAIMS_COUNT_COLS = ["Denominator", "Covid_like"]
CLAIMS_COUNT_COLS = ["Denominator", "Covid_like", "Flu1"]
CLAIMS_DATE_COL = "ServiceDate"
FIPS_COL = "fips"
DATE_COL = "timestamp"
Expand All @@ -44,6 +50,7 @@ class Config:
"PatCountyFIPS": str,
"Denominator": float,
"Covid_like": float,
"Flu1": float,
"PatAgeGroup": str,
"Pat HRR ID": str,
}
Expand Down
4 changes: 2 additions & 2 deletions claims_hosp/delphi_claims_hosp/load_data.py
Original file line number Diff line number Diff line change
Expand Up @@ -52,7 +52,7 @@ def load_claims_data(claims_filepath, dropdate, base_geo):

return claims_data

def load_data(input_filepath, dropdate, base_geo):
def load_data(input_filepath, dropdate, base_geo, numerator_name):
"""
Load in claims data, and combine them.

Expand All @@ -71,7 +71,7 @@ def load_data(input_filepath, dropdate, base_geo):

# rename numerator and denominator
data.fillna(0, inplace=True)
data["num"] = data["Covid_like"]
data["num"] = data[numerator_name]
data["den"] = data["Denominator"]
data = data[['num', 'den']]
data.reset_index(inplace=True)
Expand Down
55 changes: 31 additions & 24 deletions claims_hosp/delphi_claims_hosp/run.py
Original file line number Diff line number Diff line change
Expand Up @@ -120,30 +120,37 @@ def run_module(params):
else:
logger.info("Starting no weekday adj", geo_type=geo)

signal_name = Config.signal_weekday_name if weekday else Config.signal_name
if params["indicator"]["write_se"]:
assert params["indicator"]["obfuscated_prefix"] is not None, \
"supply obfuscated prefix in params.json"
signal_name = params["indicator"]["obfuscated_prefix"] + "_" + signal_name

logger.info("Updating signal name", signal=signal_name)
updater = ClaimsHospIndicatorUpdater(
startdate,
enddate,
dropdate,
geo,
params["indicator"]["parallel"],
weekday,
params["indicator"]["write_se"],
signal_name,
logger,
)
updater.update_indicator(
claims_file,
params["common"]["export_dir"],
)
max_dates.append(updater.output_dates[-1])
n_csv_export.append(len(updater.output_dates))

for numerator_name in ["Covid_like", "Flu1"]:
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Non-blocking: do you know when the Flu1 column became available / do you know if existence of this column is flaky?


signal_name = (
Config.signal_weekday_name[numerator_name] if weekday else Config.signal_name[numerator_name]
)
if params["indicator"]["write_se"]:
assert (
params["indicator"]["obfuscated_prefix"] is not None
), "supply obfuscated prefix in params.json"
signal_name = params["indicator"]["obfuscated_prefix"] + "_" + signal_name

logger.info("Updating signal name", signal=signal_name)
updater = ClaimsHospIndicatorUpdater(
startdate,
enddate,
dropdate,
geo,
params["indicator"]["parallel"],
weekday,
params["indicator"]["write_se"],
signal_name,
numerator_name,
logger,
)
updater.update_indicator(
claims_file,
params["common"]["export_dir"],
)
max_dates.append(updater.output_dates[-1])
n_csv_export.append(len(updater.output_dates))
logger.info("Finished updating", geo_type=geo)

# Remove all the raw files
Expand Down
16 changes: 12 additions & 4 deletions claims_hosp/delphi_claims_hosp/update_indicator.py
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,9 @@ class ClaimsHospIndicatorUpdater:
# pylint: disable=too-many-instance-attributes, too-many-arguments
# all variables are used

def __init__(self, startdate, enddate, dropdate, geo, parallel, weekday, write_se, signal_name, logger):
def __init__(
self, startdate, enddate, dropdate, geo, parallel, weekday, write_se, signal_name, numerator_name, logger
):
"""
Initialize updater for the claims-based hospitalization indicator.

Expand All @@ -45,8 +47,14 @@ def __init__(self, startdate, enddate, dropdate, geo, parallel, weekday, write_s
self.startdate, self.enddate, self.dropdate = [pd.to_datetime(t) for t in
(startdate, enddate, dropdate)]

self.geo, self.parallel, self.weekday, self.write_se, self.signal_name = \
geo.lower(), parallel, weekday, write_se, signal_name
self.geo, self.parallel, self.weekday, self.write_se, self.signal_name, self.numerator_name = (
geo.lower(),
parallel,
weekday,
write_se,
signal_name,
numerator_name,
)

# init in shift_dates, declared here for pylint
self.burnindate, self.fit_dates, self.burn_in_dates, self.output_dates = \
Expand Down Expand Up @@ -147,7 +155,7 @@ def update_indicator(self, input_filepath, outpath):

# load data
base_geo = Config.HRR_COL if self.geo == Config.HRR_COL else Config.FIPS_COL
data = load_data(input_filepath, self.dropdate, base_geo)
data = load_data(input_filepath, self.dropdate, base_geo, self.numerator_name)
data_frame = self.geo_reindex(data)

# handle if we need to adjust by weekday
Expand Down
2 changes: 1 addition & 1 deletion claims_hosp/tests/test_backfill.py
Original file line number Diff line number Diff line change
Expand Up @@ -37,7 +37,7 @@ def test_store_backfill_file(self):
backfill_df = pd.read_parquet(backfill_dir + "/"+ fn, engine='pyarrow')

selected_columns = ['time_value', 'fips', 'state_id',
'num', 'den', 'lag', 'issue_date']
'num', 'den', 'lag', 'issue_date', 'num_flu']
assert set(selected_columns) == set(backfill_df.columns)

os.remove(backfill_dir + "/" + fn)
Expand Down
4 changes: 2 additions & 2 deletions claims_hosp/tests/test_indicator.py
Original file line number Diff line number Diff line change
Expand Up @@ -20,8 +20,8 @@


class TestLoadData:
fips_data = load_data(DATA_FILEPATH, DROP_DATE, "fips")
hrr_data = load_data(DATA_FILEPATH, DROP_DATE, "hrr")
fips_data = load_data(DATA_FILEPATH, DROP_DATE, "fips", "Covid_like")
hrr_data = load_data(DATA_FILEPATH, DROP_DATE, "hrr", "Covid_like")

def test_backwards_pad(self):
num0 = np.array([0, 1, 2, 3, 4, 5, 6, 7, 8], dtype=float).reshape(-1, 1)
Expand Down
8 changes: 4 additions & 4 deletions claims_hosp/tests/test_load_data.py
Original file line number Diff line number Diff line change
Expand Up @@ -21,23 +21,23 @@
class TestLoadData:
fips_claims_data = load_claims_data(DATA_FILEPATH, DROP_DATE, "fips")
hrr_claims_data = load_claims_data(DATA_FILEPATH, DROP_DATE, "hrr")
fips_data = load_data(DATA_FILEPATH, DROP_DATE, "fips")
hrr_data = load_data(DATA_FILEPATH, DROP_DATE, "hrr")
fips_data = load_data(DATA_FILEPATH, DROP_DATE, "fips", "Covid_like")
hrr_data = load_data(DATA_FILEPATH, DROP_DATE, "hrr", "Covid_like")

def test_base_unit(self):
with pytest.raises(AssertionError):
load_claims_data(DATA_FILEPATH, DROP_DATE, "foo")

with pytest.raises(AssertionError):
load_data(DATA_FILEPATH, DROP_DATE, "foo")
load_data(DATA_FILEPATH, DROP_DATE, "foo", "Covid_like")

def test_claims_columns(self):
assert "hrr" in self.hrr_claims_data.index.names
assert "fips" in self.fips_claims_data.index.names
assert "timestamp" in self.hrr_claims_data.index.names
assert "timestamp" in self.fips_claims_data.index.names

expected_claims_columns = ["Denominator", "Covid_like"]
expected_claims_columns = ["Denominator", "Covid_like", "Flu1"]
for col in expected_claims_columns:
assert col in self.fips_claims_data.columns
assert col in self.hrr_claims_data.columns
Expand Down
8 changes: 7 additions & 1 deletion claims_hosp/tests/test_update_indicator.py
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,7 @@ def test_shift_dates(self):
self.weekday,
self.write_se,
Config.signal_name,
"Covid_like",
TEST_LOGGER
)
## Test init
Expand All @@ -74,6 +75,7 @@ def test_geo_reindex(self):
self.weekday,
self.write_se,
Config.signal_name,
"Covid_like",
TEST_LOGGER
)
updater.shift_dates()
Expand All @@ -93,6 +95,7 @@ def test_update_indicator(self):
self.weekday,
self.write_se,
Config.signal_name,
"Covid_like",
TEST_LOGGER
)

Expand All @@ -115,6 +118,7 @@ def test_write_to_csv_results(self):
self.weekday,
self.write_se,
Config.signal_name,
"Covid_like",
TEST_LOGGER
)

Expand Down Expand Up @@ -186,7 +190,7 @@ def test_write_to_csv_results(self):

def test_write_to_csv_with_se_results(self):
obfuscated_name = PARAMS["indicator"]["obfuscated_prefix"]
signal_name = obfuscated_name + "_" + Config.signal_weekday_name
signal_name = obfuscated_name + "_" + Config.signal_weekday_name["Covid_like"]
updater = ClaimsHospIndicatorUpdater(
"02-01-2020",
"06-01-2020",
Expand All @@ -196,6 +200,7 @@ def test_write_to_csv_with_se_results(self):
True,
True,
signal_name,
"Covid_like",
TEST_LOGGER
)

Expand Down Expand Up @@ -248,6 +253,7 @@ def test_write_to_csv_wrong_results(self):
self.weekday,
self.write_se,
Config.signal_name,
"Covid_like",
TEST_LOGGER
)

Expand Down