Welcome to Shanhai Xincheng | Register

Home > Industry Information > Domestic Core Running Tinymaxi's Lightweight Neural Network Reasoning Library-Mill Mill Is Based On Core Chi D9 Domestic Business Board

Domestic Core Running Tinymaxi's Lightweight Neural Network Reasoning Library-Mill Mill Is Based On Core Chi D9 Domestic Business Board

This article will introduce Mill Electronics MyD-YD9360 Shangxian Board (Mill is based on Xinchi D9360Tinymaxi lightweight neural network reasoning solution test.

 

Computing power test

 

TinyMaix It is an ultra -lightweight neural network inference library for the single -chip microcomputer, that isTo let you run a lightweight deep learning model on any single -chip machine ~

 

The constructed environment is the compiled Ubuntu18.04 that has been pre -installed with the CMAKE MAKE tool. Due to the magic network, download the TAR package to the host in advance, and then transmit it to the card to decompress it in the card card to decompress it.

 

• View cmake version

 

cmake -version

 

 

• View Make version

 

make -version

 

 

Confirm the file path, try not to copy it under the right path

 

 

Comes with examples

File structure

 

MNISTExemplary example

 

MNISTIt's handwritten digital recognition task

 

cdUse mkdir build & amp; & amp; cd built command to switch to the built folder

 

cmake ..make./ mnist

 

cmakeGenerate construction system

 

 

Use make to build executable files and run

You can see the output information

 

MNIST Examples did not use any instructions to accelerate by default. A 28 × 28 handwritten digital simulation image was run, and a total of 0.114 milliseconds were consumed.

MBNETExemplary example

 

mbnet It is a simple image classification model for mobile devices.

 

Switch to /examples /mbnet directory:

 

Modify the main.c file

 

Create a built folder and switch

 

Use the Cmake command to generate the construction system

 

Use the make command to build a system to generate executable files

 

Run executable file, the execution effect is as follows

 

 

 

MBNET Example runs an RGB image of 96 × 96 × 3, output 1000 categories, and consumes a total of 16.615 milliseconds

 

Run CIFAR10 DEMO

 

 

 

Mill Electron

 

Source: Mill Electronics

Statement on this site: This article is authorized by the author or related agencies. The purpose is to pass more information, which does not mean that this site agrees with its views, and this site does not guarantee or promise the authenticity of the content.If you need to reprint, please contact the column author. If the content of the article violates your rights and interests, please contact this site to delete in time.

Industry Information

Product Index :