docs: duo add tpu model parameter

milk-v · Jan 2, 2025 · df58e77 · df58e77
1 parent 39099eb
commit df58e77
Show file tree

Hide file tree

Showing 4 changed files with 7 additions and 15 deletions.
diff --git a/docs/duo/application-development/tpu/tpu-introduction.md b/docs/duo/application-development/tpu/tpu-introduction.md
@@ -9,14 +9,10 @@ sidebar_position: 10
 This section is only used to perform some simple tests on TPU. For TPU-related development, it is recommended to use [TDL SDK](https://milkv.io/docs/duo/application-development/tdl-sdk/tdl-sdk-introduction), which integrates most of the functions of TPU and provides some interfaces.
 :::
 
-Duo’s CPU CV1800B integrates CVITEK TPU for intelligent detection.
+The CPU of the Duo series development board integrates the CVITEK TPU for intelligent detection.
 
 TPU is an AI acceleration engine for deep learning neural networks, which can be used to accelerate image classification, object detection, face detection and recognition, segmentation, LSTM, and more. The main function of TPU is to offload CPU work and accelerate computer vision and voice-related operations.
 
-CV1800B TPU supported models:
+The model-related parameters supported by TPU are as follows:
 
-![duo](/docs/duo/tpu/duo-cv1800b-tpu-model_202307.png)
-
-:::tip
-The Duo development board is equipped with the CV1800B chip, which supports the **ONNX series** and **Caffe models**. Currently, it does not support TFLite models. In terms of quantized data types, it supports **quantization in BF16 format** and **asymmetric quantization in INT8 format**.
-:::
+<Image src='/docs/duo/tpu/duo-tpu-model.webp' maxWidth='80%' align='left' />
diff --git a/...plugin-content-docs/current/duo/application-development/tpu/tpu-introduction.md b/...plugin-content-docs/current/duo/application-development/tpu/tpu-introduction.md
@@ -9,14 +9,10 @@ sidebar_position: 10
 本章节内容仅用来对 TPU 进行一些简单的测试。对于 TPU 的相关开发，推荐使用 [TDL SDK](https://milkv.io/zh/docs/duo/application-development/tdl-sdk/tdl-sdk-introduction)，该 SDK 集成了 TPU 的大部分功能，并且提供了一些接口。
 :::
 
-Duo 的 CPU CV1800B 集成了 CVITEK TPU，用于智能检测
+Duo 系列开发板的 CPU 集成了 CVITEK TPU，用于智能检测。
 
-TPU 是深度学习神经网络的 AI 加速引擎，可用于加速图像分类、物体检测、人脸检测与识别、分割、LSTM 等。TPU 的主要功能是分担 CPU 工作， 加速计算机视觉和语音相关操作
+TPU 是深度学习神经网络的 AI 加速引擎，可用于加速图像分类、物体检测、人脸检测与识别、分割、LSTM 等。TPU 的主要功能是分担 CPU 工作， 加速计算机视觉和语音相关操作。
 
-CV1800B TPU 支持的模型
+TPU 支持的模型相关参数如下：
 
-![duo](/docs/duo/tpu/duo-cv1800b-tpu-model_202307.png)
-
-:::tip
-Duo 开发板搭载的是 CV1800B 芯片，该芯片支持 **ONNX 系列** 和 **Caffe 模型**，目前不支持 TFLite 模型。在量化数据类型方面，支持 **BF16 格式的量化** 和 **INT8 格式的非对称量化**
-:::
+<Image src='/docs/duo/tpu/duo-tpu-model.webp' maxWidth='80%' align='left' />
diff --git a/static/docs/duo/tpu/duo-cv1800b-tpu-model_202307.png b/static/docs/duo/tpu/duo-cv1800b-tpu-model_202307.png
diff --git a/static/docs/duo/tpu/duo-tpu-model.webp b/static/docs/duo/tpu/duo-tpu-model.webp