Buckets:
| #!/usr/bin/env python3 | |
| """审计 region 字符串：按来源列出取值、计数与当前映射状态。""" | |
| import pandas as pd | |
| ROOT = "/domains" | |
| mi = pd.read_parquet(f"{ROOT}/omnibrain-multi/index_timeseries.parquet") | |
| ci = pd.read_parquet(f"{ROOT}/counts-v4/counts_index.parquet") | |
| print("=" * 72) | |
| print("omnibrain-multi: region_raw × mapping_status（按 source）") | |
| print("=" * 72) | |
| cols = [c for c in ["source", "region", "region_raw", "mapping_status", "region_kind"] | |
| if c in mi.columns] | |
| print("可用列:", cols) | |
| for src, g in mi.groupby("source"): | |
| print(f"\n[{src}] rows={len(g)}") | |
| vc = g.groupby(["region_raw", "region", "mapping_status"], dropna=False).size() | |
| vc = vc.sort_values(ascending=False) | |
| for (raw, reg, st), n in vc.head(30).items(): | |
| print(f" {str(raw)[:34]:34s} -> {str(reg)[:16]:16s} [{st}] {n}") | |
| print(f" distinct region_raw = {g['region_raw'].nunique() if 'region_raw' in g else 'NA'}") | |
| print("\n" + "=" * 72) | |
| print("counts-v4: region_raw 取值") | |
| print("=" * 72) | |
| print(ci.groupby(["source", "region_raw", "region", "mapping_status"], dropna=False) | |
| .size().sort_values(ascending=False).head(40).to_string()) | |
Xet Storage Details
- Size:
- 1.18 kB
- Xet hash:
- 391ba5c56291feca06fc49048697d0f93bc6072a26ad1b5ede23075c3674c42a
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.