This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Import weight history

Hi,

I have 5 years of weight history exported from my withings scale via Nokia's health mate app. Data is in a CSV format.

I am trying to import the data to garmin connect but the import fails each time.

I have scoured the web and discovered that the csv file needs to be formatted like below, which I have copied exactly. Still no luck.

I then tried to delete 90% of the data and only import a few weeks pf data. that failed to.

Can someone at garmin please advise what the correct format is within the csv file that contains my weight history so that I can successfully upload this data.

Many thanks

Body
Date	Weight	BMI	Fat
11/05/2019	72.27	-	-
12/05/2019	72.78	-	-
18/05/2019	73.91	-	-
19/05/2019	73.40	-	-
20/05/2019	72.98	-	-
21/05/2019	72.47	-	-

Top Replies

All Replies

0 Former Member over 6 years ago

Just tried this yesterday and was surprised it worked first off... I'd try setting the BMI and Fat columns to 0 and only loading a year's worth of data at a time.

0 Former Member over 5 years ago in reply to Former Member

Hi there, can I ask how you formatted your columns?

I've manually created mine to take into account myfitnesspal weight data (which I had to be a premium member to export, which I am not).

I've created my table exactly as the example above, but still having issues importing it into Garmin Connect.

Any thoughts?

0 AllanOlesen67 over 5 years ago in reply to Former Member

You don't format columns in a .csv file. It is a pure text file where the values are separated by a comma.

So if the import doesn't work, you should open the file in a text editor, so you can see the raw text formatting and compare it to the sample .csv file you got. If you open it in a spreadsheet or anything other with fancy column formatting, you will not be able to see it the basic text format is correct.

The most common problem with CSV files is that one half of the world uses a dot as a decimal separator and the other half uses a comma.

If you are in the second half - like me - the separator between values is not a comma, because that it already used for the decimal point. In that case, a semicolon is often used to separate the values, but it is still called a .csv file, which is very misleading. So if you live in a country which uses comma as a decimal point, and you have exported the .csv file from some software which shows the data nicely in columns (for example a spreadsheet), there is a risk that this software automatically has used the "wrong" .csv format.

Next problem is date format. All kinds of pitfalls here. Both the order of year, month and day, and the separator used in the date. You really have to look carefully at the sample file and make sure you use the same format.

0 Former Member over 5 years ago in reply to Former Member

Not sure how it will come out here.. the odd bit is 2 "header lines".

Body
Date,Weight,BMI,Fat
"2019-12-11","72.201","25.5814200680272","15.51"
"2019-12-12","73.111","25.9038407029478","15.75"
"2019-12-13","72.399","25.6515731292517","15.96"
"2019-12-14","71.489","25.3291524943311","15.71"
"2019-12-16","72.535","25.6997590702948","16.13"
"2019-12-17","72.75","25.7759353741497","16.29"
"2019-12-18","73.021","25.8719529478458","16.11"

0 AllanOlesen67 over 5 years ago in reply to Former Member

This is the file, which actually works for you? In that case, there are actually several oddnesses, which may be worth copying for anyone else having problems with the import:

I agree that the first line is odd. That is probably the only time I have seen a .csv with two header lines.
The quotation marks are text qualifiers, and should actually prevent the import routine from recognizing the dates and numbers. (And even more oddly: They are not used in the header lines, which are actually text.)
European date format.

0 Former Member over 5 years ago in reply to AllanOlesen67

From memory it was duplicating an example I found. Regardless of whether it is "correct" or not, it works :-)

0 Former Member over 5 years ago in reply to Former Member

Oh yes, I believe I did reverse the date after a testing it working the other way. The import page allows you to specify and ISO is my preference for dates.

0 AllanOlesen67 over 5 years ago in reply to Former Member

Unknown said:
Regardless of whether it is "correct" or not, it works :-)

My point was not whether it was correct or not.

My point was that some of these "unusualities" may actually be the reason that it works for you - and consequently be an example worth following.

0 Former Member over 5 years ago in reply to AllanOlesen67

Just want to note that the two header lines seem necessary - I imagine that's how Fitbit exports the data, so that's what garmin expects.

I was able to import my CSV data with the following format:

body

date,weight

12/31/2019,180.6

...

0 Former Member over 5 years ago

#--OPTION: conforme DATA to CSV
#Regular expression in replace function of text editor to conforme CSV
   ^ (regular start line) replace to " (quote char)
   $ (regular end line) replace to " (quote char)
#semicolon separator option
   ; (separator character) replace to "," (quote-comma-quote char)
#TAB separator option
   \t (regular TAB) replace to "," (quote-comma-quote char)

#--header file to import
   Body
   Date,Weight,BMI,Fat,Bone,Hydration,Muscle
#GARMIN only import the four start fields
#Data under header (see date and numerical format bellow)
   "2019-10-12","90.0","25.7","18.2"
   ...
#Example
   Body
   Date,Weight,BMI,Fat,Bone,Hydration,Muscle
   "2019-12-25","100.1","30.2","25.3"

#--On GARMIN CONNECT (all languages personal parameters)
#In import function select your file and click on "Data import" button. On the message box select
   Language:English (same header terms in import file)
   Date format:yyyy-mm-dd (same format in data file)
   Number:comma or point (same format in data file)

#--Follow on www
support.garmin.com/.../
forums.garmin.com/.../import-weight-history
www.reuneker.nl/.../