Move transform logic around

This commit is contained in:
Andras Schmelczer 2026-01-31 12:48:29 +00:00
parent e1b38a1b95
commit 38b0cf1ea1
14 changed files with 1073 additions and 336 deletions

View file

@ -6,7 +6,7 @@ POSTCODE = "E14 2DG"
# Price paid: unique addresses for this postcode
pp = (
pl.scan_parquet("data_sources/pp-complete.parquet")
pl.scan_parquet("data/price-paid-complete.parquet")
.filter(pl.col("postcode") == POSTCODE)
.select("paon", "saon", "street", "postcode")
.unique()
@ -22,7 +22,7 @@ pp = (
# EPC: latest inspection per address for this postcode
epc = (
pl.scan_csv("data_sources/epc/certificates.csv")
pl.scan_csv("data/epc/certificates.csv")
.select("ADDRESS", "POSTCODE", "INSPECTION_DATE")
.filter(pl.col("POSTCODE").str.strip_chars() == POSTCODE)
.sort("INSPECTION_DATE", descending=True)