feat: add nytimes converter and example notebook

src/covid-19/covid_19_utils/covid_19_utils/__init__.py

@@ -4,4 +4,4 @@ __author__ = """Chandrasekhar Ramakrishnan"""
 __email__ = "cramakri@ethz.ch"
 __version__ = "0.1.0"
 
-from .converters import covidtracking, italy, jhu, spain, switzerland
+from .converters import covidtracking, italy, jhu, nyt, spain, switzerland

src/covid-19/covid_19_utils/covid_19_utils/converters/nyt.py

+"""
+Covid-19 converters for data from the New York Times.
+
+Data source: https://github.com/nytimes/covid-19-data
+"""
+
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+
+from . import CaseConverterImpl as CaseConverter
+from .. import helper
+
+
+class NYTCaseConverter(CaseConverter):
+    """
+    Converter for data from the U.S., collected by
+    the New York Times and hosted at https://github.com/nytimes/covid-19-data/
+    """
+
+    conversion_dict = {
+        "county": "admin2_label",
+        "state": "region_label",
+        "cases": "positive",
+        "deaths": "deceased",
+    }
+    column_list = ["date", "county", "state", "fips", "cases", "deaths"]
+
+    def convert(self, df):
+        # rename the existing columns
+        df_conv = df.rename(columns=self.conversion_dict)
+        df_conv["country"] = "USA"
+        df_conv["country_label"] = "United States of America"
+        df_conv["tested"] = np.nan
+        df_conv["date"] = pd.to_datetime(df_conv["date"])
+        # get population data
+        pop = pd.read_csv(self.atlas_folder / "us_census/co-est2019-alldata.csv")
+        pop["fips"] = pop.STATE * 1000 + pop.COUNTY
+        merged = df_conv.merge(pop[["fips", "POPESTIMATE2019"]]).rename(
+            columns={"POPESTIMATE2019": "population"}
+        )
+
+        # add NYC manually based on https://github.com/nytimes/covid-19-data#geographic-exceptions
+        nyc_county_fips = [61, 47, 81, 5, 85]
+        nyc_pop = nyc_pop = pop.loc[
+            (pop.STATE == 36) & (pop.COUNTY.isin(nyc_county_fips))
+        ]["POPESTIMATE2019"].sum()
+
+        merged = merged.append(
+            df_conv[df_conv.admin2_label == "New York City"], ignore_index=True
+        )
+        merged.loc[merged.admin2_label == "New York City", "population"] = nyc_pop
+
+        return self._set_common_columns(merged)
+
+    def read_data(self, path):
+        """Read in the county data."""
+        return pd.read_csv(Path(path) / "us-counties.csv")
+
+
+NYTCaseConverter._register()