Skip to content

Gotchas

Jump to bottom
Kannan Ramamoorthy edited this page Nov 1, 2023 · 1 revision

  • The container created for GCP VM created with marketplace colab seems to be working fine for this.

  • When there is a OOM error on consecutive fine-tuning/inferences, try restarting the container/machine. That would free-up some memory. Of course this won't work when the resource itself not sufficient for the task.

Clone this wiki locally