intraday_time_series does not correctly retrieve the data

[sefyas]

We want to obtain intraday step data and what we get is different from Fitbit dashboard. We have cases where we see zero steps for a particular date but the dashboard shows +3K steps. Or, dates where the total number of steps is different from what we see in dashboard.

This code illustrates the main steps of obtaining intra-day data and processing it (e.g. obtaining the daily counts)

# user_id, client_secret, _access_token_, _refresh_token_, start_date, and end_date are passed in as input

fitbit_auth = Fitbit(
                user_id, 
                client_secret,
                oauth2=True,
                access_token=_access_token_],
                refresh_token=_refresh_token_])

current_date = start_date;
date_increment = datetime.timedelta(days=1)
while(current_date < end_date):
   
   current_date_str = current_date.strftime("%Y-%m-%d")
   
   data = fitbit_auth.intraday_time_series(
                      'activities/steps',
                      base_date = current_date_str,
                      detail_level = '1min',
                      start_time   = '0:00',
                      end_time     = '23:59')
   
   daily_steps = 0;

   data = data['activities-steps-intraday']['dataset']
   for item in data:
       daily_steps += item['value’]  
       # other processing …

   print(f’{daily_steps} steps on {current_date_str}’)
   
   current_date += date_increment

The differences between the data we get and Fitbit dashboard do not seem to follow any pattern and appear in between dates where there is a complete match. So, I don’t think there is an issue with our calls given what we retrieve matches the dashboard some of the time; it can be an issue on the Fitbit side. Any thoughts / comment are much appreciated.

[justin-f-perez]

@sefyas This is a problem on Fitbit's side. You can prove this to yourself by using a program like Postman to make the API request directly for a date where the values don't match the dashboard, then compare the daily summary to the sum of the intraday values. Fitbit admits (somewhere on their forums a few years ago) the following conditions can cause a disparity between intraday values and daily summaries:

• User didn't sync for more than 7 days
• User's battery got too low
• User's timezone changed
• Firmware update occurred

In my experience, even a user with a charged battery that sync everyday and doesn't change timezones or update firmware can still experience this problem.

[HawkingsRainbow]

I wonder if there has been any update on this issue?

I'm unable to retrieve any intraday heartrate values, using the following code:
(Based on the great Medium tutorial by Stephen Hsu at https://towardsdatascience.com/collect-your-own-fitbit-data-with-python-ff145fa10873)

import fitbit
import gather_keys_oauth2 as Oauth2
import pandas as pd 
import datetime
CLIENT_ID = '***'
CLIENT_SECRET = '***'

server = Oauth2.OAuth2Server(CLIENT_ID, CLIENT_SECRET)
server.browser_authorize()
ACCESS_TOKEN = str(server.fitbit.client.session.token['access_token'])
REFRESH_TOKEN = str(server.fitbit.client.session.token['refresh_token'])
auth2_client = fitbit.Fitbit(CLIENT_ID, CLIENT_SECRET, oauth2=True, access_token=ACCESS_TOKEN, refresh_token=REFRESH_TOKEN)

[Browser-based authorization]

yesterday = str((datetime.datetime.now() - datetime.timedelta(days=1)).strftime("%Y%m%d"))
yesterday2 = str((datetime.datetime.now() - datetime.timedelta(days=1)).strftime("%Y-%m-%d"))
today = str(datetime.datetime.now().strftime("%Y%m%d"))

fit_statsHR = auth2_client.intraday_time_series('activities/heart', base_date=yesterday2, detail_level='1sec')

fit_statsHR contains only an 'activities-heart-rate' key. 'activities-heart-intraday' seems to be missing.

The complete contents of fit_statsHR are as follows:

In [44]: fit_statsHR
Out[44]:
{'activities-heart': [{'dateTime': '2018-10-29',
   'value': {'customHeartRateZones': [],
    'heartRateZones': [{'caloriesOut': 1316.40702,
      'max': 96,
      'min': 30,
      'minutes': 911,
      'name': 'Out of Range'},
     {'caloriesOut': 393.42948,
      'max': 135,
      'min': 96,
      'minutes': 60,
      'name': 'Fat Burn'},
     {'caloriesOut': 0,
      'max': 164,
      'min': 135,
      'minutes': 0,
      'name': 'Cardio'},
     {'caloriesOut': 0, 'max': 220, 'min': 164, 'minutes': 0, 'name': 'Peak'}],
    'restingHeartRate': 51}}]}

Setting 'base_date = "today" ' yields the same result.

At the moment, I'm testing with iPython, in a Python 3 Miniconda+Windows 10 environment.

Thanks a lot in advance for any help and/or commiseration!