Kernel Restarting Issue

Kernel Restarting or Out of Memory Error occurs when RAM chokes up :slightly_smiling_face:. We request you to select higher configuration RAM from hardware configuration settings.

The error is usually caused due to the model size or the batch size being loaded.

Here are the steps

  1. Click ‘START PROJECT’

  1. Click ‘Edit Hardware Configuration’ icon

  2. Click ‘CPU DROP-DOWN MENU’ and then select ‘16 core 60GB RAM’, Select that option

Confused which RAM to choose? :face_with_monocle:

Monitor the dynamic usage of your RAM (Total, Available, Free RAM).

Run the command in SSH/Terminal : watch -n 0.2 free -m

Additional Checks:

The following methods are best to debug the source of the error:

  • Check if the model is too big to fit on the given hardware specs. You can do this by checking the number of parameters and size of tensors and their type (float32/float64 etc.)
  • Check if multiple training sessions are running simultaneously
  • Check if some other Jupyter Notebook also is using up RAM/VRAM