on 07-30-2015 8:00 AM
Hi
I was loading data in SAP PA 2.2 automated mode, and for some reason when I click the analysis button in Data description screen the tool can't recognise the date fields as date, it will take the date column as number. And If I manually edit the type to date and run the algorithm, the tool throws error. And sometimes it does recognise the date fields. I am taking data from flat excel files.
Want to understand is there any specific format of date for which the tool recognise it correctly?
Hi,
If the date variable is not recognized when clicking "Analyze", forcing it is not helpful. It means that Automated Analytics couldn't interpret the information correctly.
To change that, you need to go 1 step back (to the data source selection panel) and click on the Advanced Settings button :
Then, click on File Settings and check the box in front of Specify Date Format. Specify from the list in which order Day, Month and Year are specified in your dataset and then enter the separator that is used as seen below.
Note that if your date variable is a string in a database and contains months names (e.g. "Jan" or "January"), it won't be possible to proceed this way. You'll need a real date variable and have to convert the string to a date or datetime.
In Excel, months may appear with their name but it's just a display option you can change so that numbers are displayed for months.
Hope this helps,
Armelle
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Though I will try the suggestions you have posted. Thanks
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hello,
I can think of a couple of causes: Excel stores the dates as numbers and then displays as dates, there might be something wrong in the way one file is read by PA. Or you might have a single value badly written in one of the files which makes PA misunderstand the content.
Can you try and export both Excel files as CSV and look into them to check if the date fields are saved differently?
For time series, check http://help.sap.com/businessobject/product_guides/pa22/en/pa22_ts_user_en.pdf P9
I am not super clear why the same formats are not currently listed on both guides. This is being investigated by our documentation team. Anyway please stick to the formats in the time series guide and check that all Excel cells are formatted fine as highlighted by Pierpaolo.
Thanks
Antoine
Hi Antoine
Thank you. I did go through the documentation. ActuallyI have two excel files with different data sets. But the date column is same in both of them are same. However for one file the tool is recognising it as date and for the other one it is not recognising!
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
In a nutshell, what you have to do is to change the format in Excel to adapt to the new format.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi,
Please refer to p.29 of http://help.sap.com/businessobject/product_guides/pa22/en/pa22_class-clust_user_en.pdf
Kind ask to you for next time :-).
Can you please make sure to start in our help portal (SAP Predictive Analytics 2.2 – SAP Help Portal Page) and search in our documentation to see if you can find insights before posting ?
Thanks & regards
Antoine
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi,
Can you please screenshot or indicate what your date format looks like in your data source?
Thanks
Antoine
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
85 | |
10 | |
10 | |
9 | |
7 | |
6 | |
5 | |
5 | |
4 | |
3 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.