![DRP-AI Translatorの特徴](/sites/default/files/media/images/drp-ai-translator-feature.png)
Overview
Description
DRP-AI Translator is a tool that generates object code for AI from trained AI models, in ONNX format that is executable on Renesas' RZ/V series of MPUs.
It optimizes the graph structure of trained AI models and FP16 quantization to minimize memory accesses and enable highly efficient and fast AI model inference processing. Optionally, you can add pre- or post-processing required for AI, such as normalization and resizing. The added pre- and post-processing is processed by DRP at high speed.
You can easily generate object code that can be implemented on an evaluation board by simply specifying an ONNX file and running the DRP-AI translator on a Linux PC.
Features
The DRP-AI translator has the four features shown in the figure below.
![DRP-AI Translatorの特徴](/sites/default/files/media/images/drp-ai-translator-feature.png)
Conversion flow
The conversion flow with the DRP-AI translator is shown in the figure below.
![DRP-AI Translatorの実装フロー](/sites/default/files/media/images/drp-ai-translator-mounting-flow.png)
Comparison
Applications
Software
Software & Tool Pages
Software title
|
Software type
|
Company
|
---|---|---|
RZ/V2L DRP-AI Support Package [V7.50] This product provides the software and documentation for DRP-AI embedded within RZ/V2L.
|
Software Package | Renesas |
RZ/V2M DRP-AI Support Package [V7.50] This product provides the software and documentation for DRP-AI embedded within RZ/V2M.
|
Software Package | Renesas |
DRP-AI Translator [V1.84] This is an AI model conversion tool (DRP-AI Translator) for DRP-AI equipped products. Please check the Release Notes and User's Manual first before using this product.
|
Software Package | Renesas |
RZ/V2MA DRP-AI Support Package[V7.50] This product provides the software and documentation for DRP-AI embedded within RZ/V2MA.
|
Software Package | Renesas |
4 items
|
Videos & Training
DRP-AI Accelerator embedded in RZ/V series MPUs provides high-speed AI processing while keeping high power efficiency at the endpoints.
News & Blog Posts
Blog Post | Apr 18, 2022 | ||
Blog Post | Feb 15, 2022 | ||
Blog Post | Nov 15, 2021 |