Thank you very much Rafo for pushing this through. We're getting close.

I'd like to strongly recommend that everyone take a look at the output and provide feedback immediately. Unless we hear otherwise this forms the basis for pass7 of the 2015 data.

Sebouh and Matt fixed the "large dPhi" issue and I have just merged in their code.

If I have some time I will look into the BorderedBandMatrix issue, but I'd like to spend my time first looking at the output.

I agree that having the reco time printed out (like it used to be) would be useful. I'm not sure when or why this was changed.


From: Hps-analysis <[log in to unmask]> on behalf of Rafayel Paremuzyan <[log in to unmask]>
Sent: Friday, November 17, 2017 4:52 AM
To: [log in to unmask]; hps-software
Subject: Re: [Hps-analysis] test pass 7d
Hi All,

new tpass7d is processed, which included
most of the fixes, except the "Large dphi"

No "Singular Matrix" exception is observed in this pass.

Instead One new exception types are observed
"BorderedBandMatrix decomposeBand not positive definite", however only
for 1 file

Also for some reason sever cached evio files had file size 0, and
consequently no reconstruction started on these files.
I did a ccpr on this.

Details are summarized in the confluence page:

ONE thing that will be highly desirable for future debugging, is to printout
time information in a Human readable format, currently
it prints 19 digit time like this below. Is anyone know how this can be
[INFO] [org.lcsim.job.EventPrintLoopAdapter] event: 30404597; time:
1431855691769105244; seq: 60000
[INFO] [org.lcsim.job.EventPrintLoopAdapter] event: 30409598; time:
1431855692068122040; seq: 65000
[INFO] [org.lcsim.job.EventPrintLoopAdapter] event: 30414597; time:
1431855692360307740; seq: 70000


On 11/14/2017 08:42 AM, Rafayel Paremuzyan wrote:
> Hi All,
> while the checklist is not fully complete yet,
> a test pass 7a called tpass7a is cooked.
> you can find relevant information in the confluence page
> NOTE: this not a final test pass before the pass7, there will be
> another test pass
> when the checklist is complete.
> This pass is reconstructed, because approvals of some items in the
> checklist require data with the newest geometry.
> Rafo

Hps-analysis mailing list
[log in to unmask]

Use REPLY-ALL to reply to list

To unsubscribe from the HPS-SOFTWARE list, click the following link: