An error occurred (ModelError) when calling the InvokeEndpoint operation: Received server error (500) from model with message "[Errno 2] No such file or directory: '/.sagemaker/mms/models/model/model.pth'".

Can someone help?

I have created a deep learning model using sagemaker using PyTorch framework. I have uploaded the model artifacts into S3 and deployed the model using the following commands.

**from sagemaker.pytorch import PyTorchModel
from sagemaker import get_execution_role
role = get_execution_role()
import boto3
import numpy as np
pytorch_model = PyTorchModel(model_data='s3://pytorchexample/model.tar.gz', role=role, entry_point='inference.py', framework_version='1.4')
predictor = pytorch_model.deploy(instance_type='ml.t2.medium', initial_instance_count=1)**

the endpoints gets created but while testing the model I get the following error:

**An error occurred (ModelError) when calling the InvokeEndpoint operation: Received server error (500) from model with message "[Errno 2] No such file or directory: '/.sagemaker/mms/models/model/model.pth'". See https://us-east-1.console.aws.amazon.com/cloudwatch/home**

I am not sure why PyTorch is looking for sagemaker folder for the model.pth file. It resides in the S3 bucket.

Start with inference.py and the model.tar.gz artifact used by PyTorchModel, then review the SageMaker deployment logs in CloudWatch for how model.pth is located. Confirm the artifact layout and inference entry point match the deployed endpoint, and verify that invoking the endpoint completes without the missing-file error.

Written by the indexing model from the issue text.