Building Robust Production-Ready Deep Learning Vision Models in Minutes, with Tensorflow Serving models can be served into production.
Is an API to use or apply with a model for inference after it has been trained, it involves having a server-client architecture and serving or exposing our trained models for inference.
In this reposiory an image classifier is developed and certain specific access patterns will be applied for the model. A brief overview of what these access patterns are:
- On the client side there is an input image which need to be classified. This image needs to be converted to an specific encoded format.
- The image must be wrapped in a specific JSON payload with headers.
- Then the image is sent to a web services/API which should typically be hosted on a server.
- The API call will invoke the pre-trained model to make the prediction and serve the inference result as a JSON response from the server to the client.
- The client gets its predictions and hopefully, it will be useful.