Create a HuggingFacePredictor from a HuggingFaceModel with entry_point

[matteoghera]

Hello,
I create my HuggingFaceModel with entry_point in this way:

from sagemaker.huggingface.model import HuggingFaceModel

huggingface_model = HuggingFaceModel(
    model_data=s3_training_model_uri,
    role=role,
    entry_point='evaluation.py',
    source_dir='./scripts',
    transformers_version='4.6',
    pytorch_version='1.7',
    py_version='py36',
    sagemaker_session=sess
)

And I try to create a HuggingFacePredictor with this statement:

predictor = huggingface_model.deploy(
    initial_instance_count=1,
    instance_type='ml.p2.xlarge'
)

The execution of the previous statement does not stop.

Can anybody help me?