Why does my kernel keep restarting when I try to download pre-trained Hugging Face BERT models weights to Amazon SageMaker?

Question

When I try to download the pre-trained Hugging Face BERT models weights to the conda_pytorch_p36 kernel of my Amazon SageMaker Notebook instance using the following command, the kernel always restarts:

```
PRE_TRAINED_MODEL_NAME2='sshleifer/distilbart-cnn-12-6'
model2 = BartForConditionalGeneration.from_pretrained(PRE_TRAINED_MODEL_NAME2, cache_dir='hf_cache_dir/')
```
Note I have installed following libraries using pip commands.

```
!pip install transformers==4.17.0
```

The result is the same for Hugging Face "facebook/bart-large-cnn" models.

Why is this happening, and how do I resolve the issue?

Answer

This typically happens when there's high resource utilization on the notebook instance and increasing instance type may help. 
Additionally, I would suggest that you open a support case under SageMaker queue by providing Sagemaker notebook ARN and associated Cloudwatch logs recorded for this notebook so that a support engineer can further troubleshoot the issue.

Why does my kernel keep restarting when I try to download pre-trained Hugging Face BERT models weights to Amazon SageMaker?

Contenido relevante