I suppose another approach might be to concatenate the description with the 
category, separated by a space, and just allow the algorithm, to learn.

> On Sep 15, 2018, at 8:27 AM, Justin H Haynes <jus...@justinhaynes.com> wrote:
> 
> One bank I use allows me to customize a set of Categories.  The categories 
> can have hierarchy denoted with “/“.  There is a facility for static mapping 
> of text found in the description and the categories.
> 
> I would like to create categories at the bank which will match the accounts 
> in GNUCash.  In this way, I can do a simple search and replace in a 
> spreadsheet program to convert “/“ to “:”, and also to prepend accounts with, 
> for example, “Liabilities:” or “Expenses:”.  The whole operation from export, 
> to search and replace, to exporting back to csv from a spreadsheet would take 
> less than a minute.
> 
> What does the “Import Transactions from CSV” wizard expect of the contents of 
> fields to which one maps “Account”?  Does it expect the full literal text 
> representing the account as I have implied in my search and replace?   For 
> example, if Date, Description and Deposit fields are defined, and the Account 
> information in the csv is “Expenses:Coffee:Equipment and Maintenance”, will 
> this be the account specified in the imported transaction?
> 
> Secondly, if the “Import Transactions from CSV” wizard does expect the 
> account in this way, will the Bayesian learning be applied to the transaction 
> so that it can benefit from the mapping and learn from the transactions?  My 
> guess is that it would not, because I would expect the transactions to behave 
> as if they were manually specified.  However, I think it would be useful to 
> allow it to learn from this import, because I will not always import 
> transactions in this way.  In other words, I will likely import a few 
> transactions every day so that I can more easily keep up with bookkeeping.  
> In this case, the algorithm can benefit from a massive import when, say, I 
> catch up my books *before* the opening date to extend them further into the 
> past and do all the reconciliation needed to make that happen.
> 
> Thanks,
> 
> Justin
> 
> 
> 

_______________________________________________
gnucash-user mailing list
gnucash-user@gnucash.org
To update your subscription preferences or to unsubscribe:
https://lists.gnucash.org/mailman/listinfo/gnucash-user
If you are using Nabble or Gmane, please see 
https://wiki.gnucash.org/wiki/Mailing_Lists for more information.
-----
Please remember to CC this list on all your replies.
You can do this by using Reply-To-List or Reply-All.

Reply via email to