What is an Inference API?
Once you have trained and evaluated a model that you are happy with on Nexus, there are a variety of ways in which you can use your model for prediction. Some of the other alternatives you might want to check out are Portal and Generating Predictions.
One such way that Datature has introduced is REST API deployment. An API is a way for you to deploy your model for inference in a way that is accessible online. Nexus now allows for hosting an API deployment of your model so you can access and receive inferences just by providing information online. To learn more about how to run your own API deployment, look at Deploying Your Trained Model as an API.
Common Questions
What are the benefits of API deployment?
API deployment on the Nexus platform allows you to have uninterrupted access to your trained model for inference without the need to reload the model or manage it on your own. This takes the responsibility of deployment off of you so that you can focus on utilising the prediction inference in the most effective manner possible. APIs also enable you to use code to automate the process of getting predictions, thus further streamlining and automating your computer vision pipeline.
How does an API function?
The trained model is hosted on assigned GPUs, and systems in Nexus allow for your inputted information to be interpreted as prediction input for the hosted model to inference upon and return an output which you can then use for your use case.
Updated about 1 year ago