在数据框python中找到时间戳之间的差异时出错

Question

I'm trying to extract timestamp data present in a dataframe and find the time difference in seconds between the time intervals. Using this time difference, I want to calculate the speed of some object (Speed = Distance/Time).
However, I'm getting a lot of errors in this time difference calculation and I'm not able to directly consider the Delta value. Can someone help me figure out what's wrong or missing in my implementation?
Thanks!

import gpxpy
import datetime
import pandas as pd

with open('test3.gpx') as fh:
    gpx_file = gpxpy.parse(fh)
    segment = gpx_file.tracks[0].segments[0]
    coords = pd.DataFrame([
    { 'time': p.time} for p in segment.points])

dist1 = 1
coords['timestamp'] = [datetime.datetime.time(d) for d in coords['time']]
timedelta = [datetime.datetime.combine(datetime.date.today(), coords.timestamp[i + 1]) - datetime.datetime.combine(datetime.date.today(), coords.timestamp[i]) for i in range(len(coords.time)-1)]
speed = dist1/timedelta

GPX File for reference: https://github.com/stevenvandorpe/testdata/blob/master/gps_coordinates/gpx/my_run_001.gpx

Error1:
This is because the denominator is of type datetime.timedelta.

 TypeError: unsupported operand type(s) for /: 'int' and 'list'

Error2:
I also tried converting the list Elements to integer, but it does not directly allow the conversion for datetime list elements

tdelta1 = [int(x) for x in timedelta]
TypeError: int() argument must be a string, a bytes-like object or a number, not 'datetime.timedelta'

Answer 1

I'd suggest using the date time datatype built into numpy and supported by Pandas, eg:

import gpxpy
import pandas as pd

with open('my_run_001.gpx') as fd:
    gpx_file = gpxpy.parse(fd)
    segment = gpx_file.tracks[0].segments[0]

    times = pd.Series([p.time for p in segment.points], name='time')

reads the values in from your file, and then you can diff and convert to second intervals easily with:

import numpy as np

dt = np.diff(times.values) / np.timedelta64(1, 's')
speed = 1 / dt