API Overview

The design philosophy of the RBLN SDK is to simplify the deployment process, allowing users to focus on developing their models and fully leveraging the benefits of RBLN NPUs with ease. We understand that dealing with a new codebase can be challenging, so we have made significant efforts to ensure that our user APIs are as straightforward and simple as possible.

The user APIs support the two most popular deep learning frameworks, TensorFlow and PyTorch. To deploy models pre-trained with TensorFlow and PyTorch, you simply need to replace the runtime part of your code with our APIs. The structure of the RBLN SDK's APIs is summarized in the diagram below.

   

The RBLN Compiler can directly compile the graph representations of TensorFlow and PyTorch, specifically tf.function for TensorFlow_v2 , GraphDef for TensorFlow_v1 , and TorchScript or torch.nn.Module for PyTorch.

The output of the compilation is: Runtime Graph, Parameter, and Compiled Data. They are saved to the disk for deployment .

The compiled output is consumed by the RBLN Runtime Runtime() object , which provides a run() method for running inference on RBLN NPU.

To gain a better understanding of the functionality of each API, we suggest that you walkthrough the PyTorch and TensorFlow tutorials.