Account associations with Description are not "remembered"

[Derek Atkins]

Hi,

Lincoln A Baxter <lab at lincolnbaxter.com> writes:

> On Tue, 2016-07-19 at 12:08 -0400, Brooks Bridges wrote:
>> Derek, you nailed my problem - for some reason I didn't even notice
>> the
>> intermediate window - so anxious to see results - damn cracks in
>> sidewalk.
>> I tried a test case just now changing in that window and all works.
>> Thanks .
>
> Derek,
>
> I think I know the answer to this because I could not find any
> references to the Imbalance accounts in my Bayes slot data, But, if one
> leaves the transactions unmatched in the import window and finishes, as
> Brooks did, does the importer store Bayesian slot data weighting the
> next imported transactions toward matching the Imbalance accounts? If
> so, I could update my gc_prune_bayes_data.pl script to explicitly
> remove any slot data referencing the Imbalance accounts. 

I *BELIEVE* that it will not "learn" from unmatched transactions.  But
I'm not 100% sure.  And if you *tell* it to go to Imbalance then it will
definitely learn that.  I'm not sure what triggers the learning,
specifically.  If might be "an account is assigned to the transaction",
which could also imply implicitly assigned accounts.

> Please remember to CC this list on all your replies.
> You can do this by using Reply-To-List or Reply-All.

-derek
-- 
       Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
       Member, MIT Student Information Processing Board  (SIPB)
       URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
       warlord at MIT.EDU                        PGP key available