Edge Software Catalog
Support for issues related to Edge Software Catalog
473 Diskussionen

tranining stops automatically in tls web ui

vkherde
Einsteiger
1.965Aufrufe

Hi Team,

I am trying to train a new model using tls docker deployment, however the training stop unexpectedly without throwing any errors. Sharing the screenshots for the same.

I am using EIS on a Virtual Machine which has specs of

RAM: 4GB, ROM: 64GB

Beschriftungen (1)
0 Kudos
1 Lösung
WengWai_C_Intel
Moderator
1.941Aufrufe

Hi @vkherde ,

For deep learning model training, it is recommended to use 32GB RAM system or above. With 4GB of RAM, most likely will hit into memory resources issue. While the deep learning model training process is running, you can use steps below to look at the log.

1) Use command 'docker ps' to list the active containers are running.

2) You should see 'dlscore' container is running. Use command 'docker logs -f dlscore' to look at the log while the model training is running.

 

In addition, it is also recommended to use a native system instead of VM for deep learning model training to reduce overheads from VM. Feel free to try the above suggestion and feedback to us. Thanks! 

Lösung in ursprünglichem Beitrag anzeigen

1 Antworten
WengWai_C_Intel
Moderator
1.942Aufrufe

Hi @vkherde ,

For deep learning model training, it is recommended to use 32GB RAM system or above. With 4GB of RAM, most likely will hit into memory resources issue. While the deep learning model training process is running, you can use steps below to look at the log.

1) Use command 'docker ps' to list the active containers are running.

2) You should see 'dlscore' container is running. Use command 'docker logs -f dlscore' to look at the log while the model training is running.

 

In addition, it is also recommended to use a native system instead of VM for deep learning model training to reduce overheads from VM. Feel free to try the above suggestion and feedback to us. Thanks! 

Antworten