Datasets:
Download code/01_cdd_shift_table.py from sleonca/sl-protostar-climate-ceviche: direct link, hf CLI and curl.
- Browser
- Download file 1.03 kB
-
https://huggingface.co/datasets/sleonca/sl-protostar-climate-ceviche/resolve/main/code/01_cdd_shift_table.py
- Command line
-
hf download hf://datasets/sleonca/sl-protostar-climate-ceviche/code/01_cdd_shift_table.py
-
curl -L -o 01_cdd_shift_table.py https://huggingface.co/datasets/sleonca/sl-protostar-climate-ceviche/resolve/main/code/01_cdd_shift_table.py
1.03 kB
| """Precompute, for every city and year up to 2018, the CDD and hottest day that year's daily | |
| temperatures would give if every day were shifted by delta °F (delta = -2 .. +8 by 0.25). | |
| Writes data/processed/shift_table.parquet. Reads only the challenge daily data (<= 2018-12-31). | |
| """ | |
| import numpy as np, pandas as pd | |
| from common import load_daily, ROOT | |
| DELTAS = np.round(np.arange(-2, 8.0001, 0.25), 2) | |
| d = load_daily() | |
| d = d.dropna(subset=["tmax", "tmin"]) | |
| tavg = ((d.tmax + d.tmin) / 2).to_numpy() | |
| key = d[["ID", "Year"]] | |
| out = key.drop_duplicates().reset_index(drop=True) | |
| g = pd.MultiIndex.from_frame(key) | |
| cols = {} | |
| for dl in DELTAS: | |
| s = pd.Series(np.maximum(0, tavg + dl - 65), index=g).groupby(level=[0, 1]).sum() | |
| cols[f"cdd_{dl:+.2f}"] = s | |
| tab = pd.DataFrame(cols).reset_index() | |
| tab["n_both"] = pd.Series(1, index=g).groupby(level=[0, 1]).sum().values | |
| p = ROOT / "data" / "processed"; p.mkdir(parents=True, exist_ok=True) | |
| tab.to_parquet(p / "shift_table.parquet", index=False) | |
| print(tab.shape, tab.Year.max()) | |