ArcGIS Pro 2.8 - Deep Learning with GPU - Fatal authorization error

10502
29
Jump to solution
05-14-2021 10:14 PM
Tim_McGinnes
Occasional Contributor III

Wanting to try out some of the new deep learning tools in Pro 2.8. I have a clean install of 2.8, cloned Python environment and the deep learning modules installed by "conda install deep-learning-essentials".

I can successfully run the deep learning tools with CPU specified in the environment variables. But when I choose GPU I  get an error box popup: "ArcGIS Pro has experienced a fatal authorization error and must close". Screenshot attached showing error and messages when running the Detect Objects tool.

I have ArcGIS Pro 2.8 Advanced and the Image Analyst extension licensed as a named user through ArcGIS Online. The same thing is occurring on two different machines, desktop with 5Gb Quadro P2000 and laptop with 8Gb GeForce RTX 2080. Both machines on Windows 10.

Am I just too quick off the mark trying to do deep learning in Pro 2.8?

29 Replies
S6
by Esri Contributor
Esri Contributor

Thanks for the update!

0 Kudos
xueyuanyao
New Contributor II

how did you solve the problem?i have the same problem,thanks

0 Kudos
S6
by Esri Contributor
Esri Contributor

hi, install patches 2.8.1 & 2.8.2

0 Kudos
xueyuanyao
New Contributor II

thanks,but i installed them,but it doesnt work,i have to use CPU to process it.do you know why?

0 Kudos
S6
by Esri Contributor
Esri Contributor

Are you getting the same GPU authorization error?

Are you using Named-user license or concurrent license?

If possible, share some screenshots.

0 Kudos
Shingo-Ikeda
New Contributor III

Could you specify the details of the bug that was fixed for the Classify Objects. I am running into some issues, but I am not sure if they are related.

Shingo Ikeda
Geospatial Data Scientist/Developer - Geographical Information Platform
Global Power Generation - Digital Satellite USA and Canada
0 Kudos
samaloysius4
New Contributor III

Hello Shingo-Ikeda, 

The Classify Objects tool was taking a long time to run for larger datasets. . The inference process is intensive but it should not take you many hours for a couple of thousand features. This has been fixed in 2.8.2.  We are also looking into fixing an issue with the incorrect progress bar where it shows 100% complete when the tool is still running as well as varying results on different extent runs. That will be fixed shortly. If you are facing other issues please tell us your issue and we will look into it . Thanks.

RobHodder
New Contributor II

Thanks for the information. I have been running into the exact issue with Classify Objects using DL. Looking forward to trying 2.8.2!

0 Kudos
S6
by Esri Contributor
Esri Contributor

Hi,

I am facing progress bar issues in 2.8.2 as well (Detect objects using deep learning). 

 

0 Kudos
SolanaFoo2
Occasional Contributor

I'm in 2.9.2 and when I choose GPU instead of CPU it constantly errors out!

0 Kudos