332. TorchServe

Deploying Your Model

TorchServe allows you to expose a WEB API for your Pytorch model that may be accessed directly or via your application.

3 Steps

1. Choose a default handler or author a custom model handler.
   You will define a class including preprocessing/inference/postprocessing functions.
2. Create a “model archive(all the files you need for the model to run inference)” and place it in the model store.
3. Start TorchServe and load a model archive for serving via the inference API
4. Configure and control TorchServe through the Management API.

If you are interested, please check out the Demo too.