如何在python中获得持续时间的平均值

Question

I was trying to find the answer, but nothing works.

This is my data frame:

Week_summary_df

and I need to get mean value of each activity per week.

This is my code:

from tkinter import *
from datetime import datetime, date
import csv
import pandas as pd

main_window = Tk()
main_window.title('')

def button_bau():
    current_date = date.today()
    now = datetime.now()
    current_time = now.strftime("%H:%M:%S")
    comment = entry_comment.get()

    data = {'Date': current_date, 'Time': current_time, 'Action': 'BAU', 'Comment': comment}
    with open('timesheet.csv', 'a') as csvFile:
        writer = csv.DictWriter(csvFile, data.keys())
        if csvFile.tell() == 0:
            writer.writeheader()
        writer.writerow(data)
        print(data)
    entry_comment.delete(0, END)

def button_training():
    current_date = date.today()
    now = datetime.now()
    current_time = now.strftime("%H:%M:%S")
    comment = entry_comment.get()

    data = {'Date': current_date, 'Time': current_time, 'Action': 'Training', 'Comment': comment}
    with open('timesheet.csv', 'a') as csvFile:
        writer = csv.DictWriter(csvFile, data.keys())
        if csvFile.tell() == 0:
            writer.writeheader()
        writer.writerow(data)
        print(data)
    entry_comment.delete(0, END)

def button_meeting():
    current_date = date.today()
    now = datetime.now()
    current_time = now.strftime("%H:%M:%S")
    comment = entry_comment.get()


    data = {'Date': current_date, 'Time': current_time, 'Action': 'Meeting', 'Comment': comment}
    with open('timesheet.csv', 'a') as csvFile:
        writer = csv.DictWriter(csvFile, data.keys())
        if csvFile.tell() == 0:
            writer.writeheader()
        writer.writerow(data)
        print(data)
    entry_comment.delete(0, END)

def button_break():
    current_date = date.today()
    now = datetime.now()
    current_time = now.strftime("%H:%M:%S")
    comment = entry_comment.get()

    data = {'Date': current_date, 'Time': current_time, 'Action': 'Break', 'Comment': comment}
    with open('timesheet.csv', 'a') as csvFile:
        writer = csv.DictWriter(csvFile, data.keys())
        if csvFile.tell() == 0:
            writer.writeheader()
        writer.writerow(data)
        print(data)
    entry_comment.delete(0, END)

def button_comment():
    current_date = date.today()
    now = datetime.now()
    current_time = now.strftime("%H:%M:%S")
    comment = entry_comment.get()

    data = {'Date': current_date, 'Time': current_time, 'Action': 'EOD', 'Comment': comment}
    with open('timesheet.csv', 'a') as csvFile:
        writer = csv.DictWriter(csvFile, data.keys())
        if csvFile.tell() == 0:
            writer.writeheader()
        writer.writerow(data)
        print(data)
    entry_comment.delete(0, END)

def myFormatString(bigString):
    return bigString.replace("0 days ", "").replace(".000000000", "")

def export_to_excel():

    today_df = pd.read_csv('timesheet.csv')
    today_df["full start date"] = pd.to_datetime(today_df["Date"] + " " + today_df["Time"])

    end_times_list = today_df["Time"].to_list()[1:]
    today_df["end time"] = pd.Series(end_times_list)

    end_dates_list = today_df["Date"].to_list()[1:]
    today_df["end date"] = pd.Series(end_dates_list)

    today_df["full end date"] = pd.to_datetime(today_df["end date"] + " " + today_df["end time"])
    today_df["Duration:  hh:mm:ss"] = pd.to_timedelta(today_df["full end date"] - today_df["full start date"])
    today_df["week"] = today_df["full start date"].dt.week

    day_data_df = today_df[["Date", "Action", "Duration:  hh:mm:ss"]]
    day_summary_df = day_data_df.groupby(["Date", "Action"]).sum().astype(str).applymap(myFormatString).unstack(level=-1)

    today_df["Duration:  hh:mm:ss"]=today_df["Duration:  hh:mm:ss"]
    week_data_df = today_df[["week", "Action", "Duration:  hh:mm:ss"]]
    week_summary_df = week_data_df.groupby(["week", "Action"]).sum().astype(str).applymap(myFormatString).unstack(level=-1)

    print(week_summary_df)

    excel_filename = 'timesheet.xlsx'
    writer = pd.ExcelWriter(excel_filename, engine='xlsxwriter',
                        datetime_format='hh:mm:ss')

    day_summary_df.drop(day_summary_df.columns[2], axis=1, inplace=True)
    day_summary_df.to_excel(excel_filename)
    day_summary_df.to_excel(writer, sheet_name='Daily Summary')

    week_summary_df.drop(week_summary_df.columns[2], axis=1, inplace=True)
    week_summary_df.to_excel(excel_filename)
    week_summary_df.to_excel(writer, sheet_name='Weekly Summary')

    writer.save()

l1 = Label(main_window, text='Timesheet Collector')
l1.grid(row=0, column=1, pady=5)

b1 = Button(main_window, text='BAU', width=8, bg='#CC8899', command=button_bau)
b1.grid(row=1, column=1, pady= 5, padx=60)

b2 = Button(main_window, text='Training', width=8, bg='#CC8899', command=button_training)
b2.grid(row=2, column=1, pady=5, padx=60)

b3 = Button(main_window, text="Meeting", width=8, bg='#CC8899', command=button_meeting)
b3.grid(row=3, column=1, pady=5, padx=60)

b4 = Button(main_window, text="Break", width=8, bg='#CC8899', command=button_break)
b4.grid(row=4, column=1, pady=5, padx=60)

b5 = Button(main_window, text="EOD", width=8, bg='#E0115F', command=button_comment)
b5.grid(row=5, column=1, pady=5, padx=60)

entry_comment = Entry(main_window, width=25)
entry_comment.grid(row= 6, column=1, pady=10)

b6 = Button(main_window, text="Export to Excel", width=11, bg='#C0C0C0', command=export_to_excel)
b6.grid(row=7, column=1, pady=5, padx=60)

main_window.mainloop()

do you have any idea how can I fix it?

Please note that I am a complete beginner and I got a task to create tool which will be gathered information about analysts activities during the day. And this tool needs to first gathered info to csv and then export to excel two kind of information: 1. sum of all activities during the day - I was able to manage 2. average duration of each activity during the week - for example: analyst X during the week was on meeting 30min average

Answer 1

Try this:

for col in today_df :
    print(df[col].apply(lambda x: x.seconds).groupby([pd.Grouper(freq='W-MON')]).mean().apply(lambda x: timedelta(seconds=x)))