假设表格名称为record,包含三列:ID,date1,date2
在SQL中可以这样写:
SELECT date1, date2, COUNT(DISTINCT ID) FROM record GROUP BY date1, date2;
在Python中可以使用pandas库:
import pandas as pd
record = pd.read_csv("record.csv", dtype={"ID": str})
record["date1"] = pd.to_datetime(record["date1"]) record["date2"] = pd.to_datetime(record["date2"])
result = record.groupby(["date1", "date2"]).agg({"ID": pd.Series.nunique})
print(result)