👀 Views: 74 💬 Answers: 1 📅 Created: 2025-06-12

oci tensorflow model-deployment data-science Python

I'm confused about I'm stuck on something that should probably be simple... I'm trying to deploy a TensorFlow model using OCI Data Science, but I'm getting an 'InvalidParameter' behavior when I run the deployment script. The behavior message states `InvalidParameter: The parameter 'modelPath' is invalid. Please ensure it's a valid model path.` Here's the code snippet I've been using: ```python import oci from oci.data_science import DataScienceClient config = oci.config.from_file() client = DataScienceClient(config) model_deployment_details = { 'model_id': 'ocid1.model.oc1..exampleuniqueID', 'model_path': 'oci://my_bucket/my_model_directory/', 'deployment_name': 'my_model_deployment', } response = client.create_model_deployment(model_deployment_details) ``` I've double-checked that the model path is correct, pointing to an OCI Object Storage bucket where my model files are stored. I'm using TensorFlow version 2.6, and my model was saved correctly with `tf.saved_model.save()`. I also verified that the OCI Data Science service has the necessary permissions to access this bucket. I tried changing the path format and ensuring that all necessary files are in the specified directory, but nothing seems to resolve the scenario. Has anyone encountered this question before or have suggestions on how to properly set up the model path for deployment? This is part of a larger CLI tool I'm building. What's the best practice here? My development environment is Ubuntu 20.04. Any pointers in the right direction?