Hi everyone, I have a strange problem with running transformer models on Google Cloud VM instances. Once training should start, the program is stuck without an error message at ***** Running training ***** Num examples = 6838 Num Epochs = 5 Instantaneous batch size per device = 8 Total train …

I am running into a similar issue (also on GCP). Even though I am using a different script . Training is stuck immediately. Did you find any solution?

As agademic wrote, uninstalling torch_xla did the trick for me.

Transformers on GCP Training stuck on start

🤗Transformers

agademic March 17, 2023, 4:35pm 3

This was the solution for me.

Topic		Replies	Views
Trainer.train stalls when using tutorial example on GCP instance Beginners	2	940	February 15, 2023
Training of GPT hang during Checkpoint stage 🤗Transformers	0	153	January 23, 2024
Trainer.train() is stuck 🤗Transformers	5	7749	May 1, 2023
Troubleshooting 🤗Transformers	0	244	March 23, 2023
Trainer object high memory usage on Google Cloud Platform Workbench instance 🤗Transformers	0	49	September 16, 2024

Transformers on GCP Training stuck on start

Related topics