def main():
# Load data
df = pd.read_csv(DATA_URL)
# Check 1
cols = ["datum", "vakcina", "kraj_nuts_kod", "kraj_nazev", "vekova_skupina", "prvnich_davek", "druhych_davek", "celkem_davek"]
if not all([col in df.columns for col in cols]):
raise Exception("API changed")
# Column renaming
df = df.rename(columns={
"datum": "date",
"kraj_nazev": "region",
"prvnich_davek": "people_vaccinated",
"druhych_davek": "people_fully_vaccinated",
"celkem_davek": "total_vaccinations"
})
# Add counts per day
df = df.groupby(by=["date", "region"]).agg(
people_vaccinated=("people_vaccinated", sum),
people_fully_vaccinated=("people_fully_vaccinated", sum),
total_vaccinations=("total_vaccinations", sum)
).reset_index()
# Check 2
if not (df["total_vaccinations"] == df["people_vaccinated"] + df["people_fully_vaccinated"]).all():
raise Exception("Error in columns. dose_1 + dose_2 != total_doses")
# Rename regions
df.loc[:, "region"] = df.loc[:, "region"].replace(REGION_RENAMING)
df.loc[:, "location"] = COUNTRY
# ISO
df = ISODB().merge(df, country_iso=COUNTRY_ISO)
# Compute cumsums
df = df.sort_values(by="date")
df.loc[:, "total_vaccinations"] = df.groupby("region")["total_vaccinations"].cumsum().values
df.loc[:, "people_vaccinated"] = df.groupby("region")["people_vaccinated"].cumsum().values
df.loc[:, "people_fully_vaccinated"] = df.groupby("region")["people_fully_vaccinated"].cumsum().values
# Export
export_data(
df=df,
data_url_reference=DATA_URL_REFERENCE,
output_file=OUTPUT_FILE
)
def _postprocess(self, df):
df = ISODB().merge(df, country_iso=self.country_iso, mode=self.mode_iso_merge)
df = df.sort_values(by="date")
for field in self.do_cumsum_fields:
df[field] = df.groupby("region")[field].cumsum().values
# TODO: Insert here population info (need path to population.csv as class attribute)
return df