I am trying to create a new column "Starting_time" by subtracting 60 days out of "Harvest_date" but I get the same date each time. Can someone point out what did I do wrong please?

import pandas as pd
from datetime import timedelta


df1 = pd.read_csv (r'C:\Flower_weight.csv')

def subtract_days_from_date(date, days):

    subtracted_date = pd.to_datetime(date) - timedelta(days=days)
    subtracted_date = subtracted_date.strftime("%Y-%m-%d")

    return subtracted_date

df1['Harvest_date'] = pd.to_datetime(df1.Harvest_date)
df1.style.format({"Harvest_date": lambda t: t.strftime("%Y-%m-%d")})

for harvest_date in df1['Harvest_date']:
    df1["Starting_date"]=subtract_days_from_date(harvest_date,60)

print(df1["Starting_date"])

CodePudding user response：

I am not sure if the use of the loop was necessary here. Perhaps try the following:

df1_dates['Starting_date'] = df1_dates['Harvest_date'].apply(lambda x: pd.to_datetime(x) - timedelta(days=60))
df1_dates['Starting_date'].dt.strftime("%Y-%m-%d")
df1_dates['Starting_date']

CodePudding user response：

You're overwriting the series on each iteration of the last loop

for harvest_date in df1['Harvest_date']:
    df1["Starting_date"]=subtract_days_from_date(harvest_date,60)

You can do away with the loop by vectorizing the subtract_days_from_date function. You could also reference an index with enumerate

np.vectorize

import numpy as np


subtract_days_from_date = np.vectorize(subtract_days_from_date)

df1["Starting_date"]=subtract_days_from_date(df1["Harvest_date"], 60)

enumerate

for idx, harvest_date in enumerate(df1['Harvest_date']):
    df1.iloc[idx][ "Starting_date"]=subtract_days_from_date(harvest_date,60)