Closed danielsoy closed 6 months ago
Hi Daniel. It might be some visual studio incompatibility. Options to test:
Thanks Huseyn, I got access after pulling the image, from intelecai/automl_server
Choosing this option: Double click on the start_servers.bat file in the windows folder.
Now I want to experiment with some model, appropiate for anomaly detection.
I have watched your youtube video:
https://www.youtube.com/watch?v=Ra3hYQMs1dc&ab_channel=IntelecAI
And I would like try something similar.
Ok, good luck :)
On Fri, Nov 26, 2021, 20:43 danielsoy @.***> wrote:
Thanks Huseyn, I got access after pulling the image, from intelecai/automl_server
Choosing this option: Double click on the start_servers.bat file in the windows folder.
Now I want to experiment with some model, appropiate for anomaly detection.
I have watched your youtube video:
https://www.youtube.com/watch?v=Ra3hYQMs1dc&ab_channel=IntelecAI
And I would like try something similar.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/intelec-ai/intelecai-installation/issues/1#issuecomment-980358977, or unsubscribe https://github.com/notifications/unsubscribe-auth/AM3YSUCEEBTYVRVPKC7E3RTUN7PNXANCNFSM5IZCGJRA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
Almost there...
When I click on the START button to start trainning, asks for confirmation, I click on YES and STATUS changes to: running, the percentage stays in 0%, and a couple of seconds after, STATUS gets back to : not running.
You need to check the docker logs of intelec-backend service
Get the ID of the docker service intelec-backend by running "docker service ls"
Run "docker service logs [serviceID]"
One possible cause of your problem is that your dataset is not in correct format.
On Fri, Nov 26, 2021, 23:19 danielsoy @.***> wrote:
Almost there...
When I click on the START button to star trainning, asks for confirmation, i click on YES and STATUS changes to: running, the percentage stays in 0%, and a couple of seconds after, STATUS gets back to : not running.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/intelec-ai/intelecai-installation/issues/1#issuecomment-980453764, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA5NYENJ34F7UJ2CZHJ4LV3UOABXRANCNFSM5IZCGJRA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
yes, maybe the dataset is to blame.
I will search for another short dataset, known for working well.
Do I need a CUDA GPU, for trainning by the way?
No, you dont have to have CUDA for training.
On Sat, Nov 27, 2021, 02:02 danielsoy @.***> wrote:
Do I need a CUDA GPU, for trainning by the way?
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/intelec-ai/intelecai-installation/issues/1#issuecomment-980479196, or unsubscribe https://github.com/notifications/unsubscribe-auth/AM3YSUFT6YSFFI3Y4VDEABDUOAUZJANCNFSM5IZCGJRA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
I cant find anything wrong with my log... I guess.
You need to click on the "Start" button on the training page to let the training start and stop. Then you can get the docker log of intelec_backend service.
I got : Could not do one pass in your dataloader, there is something wrong in it at some point inside the log file
I have chosen to put my two classes data, inside two separated folders, without a .cvs file, for labels(option 1). Is that correct? or a csv with labes is always needed, despite having two folders, one for each class? .****
You can watch this video to learn about how you can train an image classifier. There are also videos about object detection and image segmentation.
Please also check that your image folders contain only images, no text files, no hidden files (.DS_Store, .Trash, .git, etc).
Hello Huseyn.
Congratulations for your development, I hope you are doing well.
I was trying to run start_server.bat from visual studio code, via docker, without success.
Please look: https://www.youtube.com/watch?v=HTCVs38wDpE&ab_channel=tektronix475 https://www.youtube.com/watch?v=qcmOfih7zSg&ab_channel=tektronix475
Am I doing something wrong?
Thank you so much!