publicdata/j40-cejst-2
1
0
Fork 0
mirror of https://github.com/DOI-DO/j40-cejst-2.git synced 2025-07-28 23:11:16 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

Analysis by region (#385)

Browse source 
* Adding regional comparisons

* Small ETL fixes
This commit is contained in:
Lucas Merrill Brown 2021-07-26 08:02:25 -07:00 • committed by GitHub
commit 67b39475f7
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
7 changed files with 440 additions and 158 deletions
Download patch file Download diff file Expand all files Collapse all files

16
data/data-pipeline/etl/sources/census/etl_utils.py
View file

@ -1,5 +1,6 @@
from pathlib import Path
import csv
import pandas as pd
import os
from config import settings
@ -53,3 +54,18 @@ def get_state_fips_codes(data_path: Path) -> list:
fips = row[0].strip()
fips_state_list.append(fips)
return fips_state_list
def get_state_information(data_path: Path) -> pd.DataFrame:
"""Load the full state file as a dataframe.
Useful because of the state regional information.
"""
fips_csv_path = data_path / "census" / "csv" / "fips_states_2010.csv"
df = pd.read_csv(fips_csv_path)
# Left pad the FIPS codes with 0s
df["fips"] = df["fips"].astype(str).apply(lambda x: x.zfill(2))
return df