Advanced search

Message boards : News : [wildlife] new apps, new workunits

Author Message
Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 6592 - Posted: 31 Dec 2016, 4:00:54 UTC

I've updated the apps for all operating systems to version 0.08. Let me know if you're having any issues with these.

If these work I have most of things in place to get the assimilator working and start a constant stream of workunits. Hopefully validation works well on these. Also, I believe the checkpointing issue for the linux app should be working now. If not please let me know.

Profile JumpinJohnny
Avatar
Send message
Joined: 24 Sep 13
Posts: 237
Combined Credit: 10,275,610
DNA@Home: 192,548
SubsetSum@Home: 201,740
Wildlife@Home: 9,881,323
Wildlife@Home Watched: 55,997,833s
Wildlife@Home Events: 15,584
Climate Tweets: 333
Images Observed: 351

              
Message 6593 - Posted: 31 Dec 2016, 5:15:02 UTC - in response to Message 6592.

On my windoz box
EXACT Convolutional Neural Network Trainer v0.08
>>took each under 10 minutes
>>started validating
>>'paid' about the same credits/CPU sec as the version 7

Happy New Years!
____________

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 6594 - Posted: 31 Dec 2016, 5:33:23 UTC - in response to Message 6593.

On my windoz box
EXACT Convolutional Neural Network Trainer v0.08
>>took each under 10 minutes
>>started validating
>>'paid' about the same credits/CPU sec as the version 7

Happy New Years!


Yeah, i made a little mistake where I had the application only work on a subset of the training images.... sooooo everyone enjoy the extra credit for these WUs. :)

Profile JumpinJohnny
Avatar
Send message
Joined: 24 Sep 13
Posts: 237
Combined Credit: 10,275,610
DNA@Home: 192,548
SubsetSum@Home: 201,740
Wildlife@Home: 9,881,323
Wildlife@Home Watched: 55,997,833s
Wildlife@Home Events: 15,584
Climate Tweets: 333
Images Observed: 351

              
Message 6595 - Posted: 31 Dec 2016, 6:41:31 UTC

linux wingmen do not appear to be validating
a closer look at a linux box reveals problems
http://csgrid.org/csg/results.php?hostid=29310
13 errors, no valids

Profile JumpinJohnny
Avatar
Send message
Joined: 24 Sep 13
Posts: 237
Combined Credit: 10,275,610
DNA@Home: 192,548
SubsetSum@Home: 201,740
Wildlife@Home: 9,881,323
Wildlife@Home Watched: 55,997,833s
Wildlife@Home Events: 15,584
Climate Tweets: 333
Images Observed: 351

              
Message 6596 - Posted: 31 Dec 2016, 6:43:51 UTC - in response to Message 6594.
Last modified: 31 Dec 2016, 6:44:44 UTC

On my windoz box
EXACT Convolutional Neural Network Trainer v0.08
>>'paid' about the same credits/CPU sec as the version 7


Yeah, i made a little mistake where I had the application only work on a subset of the training images.... sooooo everyone enjoy the extra credit for these WUs. :)


No, you misunderstood. The credits are the same. "per cpu seconds"
Still quite low.

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 6597 - Posted: 31 Dec 2016, 17:07:43 UTC - in response to Message 6596.

On my windoz box
EXACT Convolutional Neural Network Trainer v0.08
>>'paid' about the same credits/CPU sec as the version 7


Yeah, i made a little mistake where I had the application only work on a subset of the training images.... sooooo everyone enjoy the extra credit for these WUs. :)


No, you misunderstood. The credits are the same. "per cpu seconds"
Still quite low.


Looking into it right now. Have a new app version which should fix the missing library issue for linux (hopefully).

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 6598 - Posted: 31 Dec 2016, 17:27:38 UTC - in response to Message 6596.

On my windoz box
EXACT Convolutional Neural Network Trainer v0.08
>>'paid' about the same credits/CPU sec as the version 7


Yeah, i made a little mistake where I had the application only work on a subset of the training images.... sooooo everyone enjoy the extra credit for these WUs. :)


No, you misunderstood. The credits are the same. "per cpu seconds"
Still quite low.



Ahh, figured out the credit issue. Validator needed the --credit_from_wu flag.


Post to thread

Message boards : News : [wildlife] new apps, new workunits