Error in Week 2 - Lab - AWS Sagemaker Notebook keeps restarting

I have run all the cells in the Jupyter Notebook as provided, but whenever I try to run the cell with trainer.train() and peft_trainer.train() , the notebook keeps restarting. I’m unable to practice or even look at the later code results.
I have not changed any parameters that might increase the load on the system. I have done “Run all cells” too.

Can someone please help.? Thank you

2 Likes

I have the same issue here, did you find out how to solve this ?

2 Likes

Not yet. Still looking for help. My assignment is due to today i.e., Oct 15, 2023.

2 Likes

I have same issue and today is my due date . can some one please help

2 Likes

try to make sure the instance setting aligned with the instruction in the lab, this worked for me.

1 Like

Tried that too. My assignment is overdue now. Not sure what to do

1 Like

Hi,

Your screenshot shows that you have an instance with just 2vCPU + 4GB. It is probably too weak for training.

The recommended instance is the ml.m5.2xlarge, giving you 8vCPU + 32GB. Try that and see if it works.

1 Like

That worked!!!

1 Like