Giter VIP home page Giter VIP logo

Comments (16)

Linzaer avatar Linzaer commented on May 22, 2024

@Dreamgang ,ncnn目前还没量化测试过,有空测试下。这几天加入人脸五点关键点,加入后一并量化测试下。

from ultra-light-fast-generic-face-detector-1mb.

zxb-caffe avatar zxb-caffe commented on May 22, 2024

The same is true of my problem. After quantification, there is an error. Excuse me, have you solved it?

from ultra-light-fast-generic-face-detector-1mb.

zxb-caffe avatar zxb-caffe commented on May 22, 2024

你好,我用ncnn提供的量化工具,将您给的模型ncnn.bin.ncnn.param进行量化,再用量化后的模型,用ncnn调用预测,结果出错了,请问您能否提供一下量化后的模型,谢谢!

我也遇到相同的问题,请问您解决了吗?

from ultra-light-fast-generic-face-detector-1mb.

Linzaer avatar Linzaer commented on May 22, 2024

@zxb-caffe MNN量化模型 ,测试正常。

from ultra-light-fast-generic-face-detector-1mb.

zxb-caffe avatar zxb-caffe commented on May 22, 2024

请问:int8和float32的模型运行时间有区别吗?我感觉几乎没变化。

from ultra-light-fast-generic-face-detector-1mb.

Linzaer avatar Linzaer commented on May 22, 2024

在arm上效果蛮明显的,特别是多核。

from ultra-light-fast-generic-face-detector-1mb.

Linzaer avatar Linzaer commented on May 22, 2024

刚开始的几次检测不会很稳定,时间会有波动,后面就逐渐稳定了。你可以for循环推理部分测一下。

from ultra-light-fast-generic-face-detector-1mb.

zxb-caffe avatar zxb-caffe commented on May 22, 2024

我是在笔记本上跑的,几乎没什么效果。您知道是为什么吗?

from ultra-light-fast-generic-face-detector-1mb.

Linzaer avatar Linzaer commented on May 22, 2024

MNN和NCNN这种移动端推理框架一般不针对PC架构处理器优化的,所以int8反而会比fp32慢的。

from ultra-light-fast-generic-face-detector-1mb.

Linzaer avatar Linzaer commented on May 22, 2024

树莓派4B MNN 320x240 ,1~4核推理时间(ms) FP32为38.57/26.35/22.66/19.68, int8为35.56/19.63/14.5/10.8, 加速效果还是很明显的。

from ultra-light-fast-generic-face-detector-1mb.

zxb-caffe avatar zxb-caffe commented on May 22, 2024

树莓派4B是只有cpu吗?没有带GPU吗?

from ultra-light-fast-generic-face-detector-1mb.

Linzaer avatar Linzaer commented on May 22, 2024

有,不过只是用来处理视频编解码的,不能用来做通用加速。

from ultra-light-fast-generic-face-detector-1mb.

zxb-caffe avatar zxb-caffe commented on May 22, 2024

好的,非常感谢。

from ultra-light-fast-generic-face-detector-1mb.

zxb-caffe avatar zxb-caffe commented on May 22, 2024

请问,量化后的准确度是怎么样的?

from ultra-light-fast-generic-face-detector-1mb.

Linzaer avatar Linzaer commented on May 22, 2024

我测试了一些图,检测结果几乎没区别。有空我单独测试下widerface的验证集结果看看区别,差别应该不大。

from ultra-light-fast-generic-face-detector-1mb.

zxb-caffe avatar zxb-caffe commented on May 22, 2024

好的。谢谢。

from ultra-light-fast-generic-face-detector-1mb.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.