search

openai / shap-e

Generate 3D objects conditioned on text or images
MIT License
11.54k stars 928 forks source link

Deploy Shap-E model in Azure Machine Learning as online endpoint #99

Open eavidal opened 1 year ago

eavidal commented 1 year ago

I'm looking to deploy the Shap-E model in Azure Machine Learning as an online endpoint, but everything I've tried so far didn't work. Did anyone by any chance get something similar working?

eliecer696 commented 1 year ago

To deploy the Shap-E model in Azure Machine Learning as an online endpoint, you can use Azure ML managed online endpoints. These provide a powerful solution for deploying machine learning models to production. With built-in security, scalability, and reliability, you can confidently deploy models as accessible web APIs for clients1.

You can create a managed online endpoint from the Models page in the studio. This is an easy way to add a model to an existing managed online deployment2.

You can also learn how to use an online endpoint to deploy your model by following this tutorial3.

nebomuk commented 1 year ago

I got something similar working although I used another service this time (vast). You need to select (or create) a container on Azure where cuda/pytorch is already set up. Then simply clone this repo, run pip install -e . and everything should work.