Pandas: Remove Year-Week rows before today?

Question

I want to drop rows before current week from my dataframe. The intended code is not working though, as the single digit weeks are still showing up. Is there a better way?

import pandas as pd
import numpy as np
from datetime import date, datetime, timedelta

data = {
    "Year": [2019, 2020, 2020, 2020, 2020, 2020, 2020],
    "Week": [40, 8, 9, 10, 11, 12, 13]
}
df = pd.DataFrame(data)

# Current YearWeek
year_week = datetime.now().strftime("%Y/W%V")
print(year_week)

df["Year/Week"] = pd.to_datetime(
    (df["Year"].astype(str) + "/W" + df["Week"].astype(str)),
    format="%Y/W%V",
    errors="ignore")

# Drop rows that have Year-Week value less than current Year-Week
df["Exclude Rows"] = np.where(
    pd.to_datetime(
        (df["Year"].astype(str) + "/W" + df["Week"].astype(str)),
        format="%Y/W%V",
        errors="ignore",
    ) < year_week, "Yes", "No")

# Drop rows
df.drop(df.loc[df["Exclude Rows"] == "Yes"].index, inplace=True)

print(df)

The output I am getting:

   Year  Week Year/Week Exclude Rows
1  2020     8   2020/W8           No
2  2020     9   2020/W9           No
5  2020    12  2020/W12           No
6  2020    13  2020/W13           No

score 0 · Answer 1 · answered Mar 19 '20 at 03:41

Here's a possible solution, inspired by this answer:

import pandas as pd
from datetime import datetime

data = {
    "Year": [2019, 2020, 2020, 2020, 2020, 2020, 2020],
    "Week": [40, 8, 9, 10, 11, 12, 13]
}
df = pd.DataFrame(data)

df = df[pd.to_datetime(df.Year.astype(str), format='%Y') + \
    pd.to_timedelta(df.Week.mul(7).astype(str) + ' days')
    > datetime.now()
]

Result:

   Year  Week
5  2020    12
6  2020    13

Pandas: Remove Year-Week rows before today?

1 Answers1