Giter VIP home page Giter VIP logo

Comments (4)

lsh1994 avatar lsh1994 commented on June 9, 2024

一个比较常见的错误是, 预测结果是标签值(0,1,2),直接(在看图应用上)看图(0-255)就是黑色的嗷,或许plt可视化效果好些。关于这个问题,请检查模型额输入和输出,以及是否为对应类别数量等

from keras-segmentation.

Niccccci avatar Niccccci commented on June 9, 2024

感谢您的解答,我的预测结果不是黑色的,是彩色的,但是感觉没有正常预测图像。我还有一个疑问跟您请教一下,一般对图像分割的效果衡量还有一个参数是meanIOU,您这边的这个acc指的是每个像素点预测的准确率吗?另外在train文件中的类别数量是11,但是我打印您的label的索引值数量是0-11,这样的话类别数量不应该是12吗?在Unet模型文件中的类别和输入图像长宽类别也要根据自己的数据集改变吧?

from keras-segmentation.

lsh1994 avatar lsh1994 commented on June 9, 2024

是的。
visualizeDataset.py中类别设为了11,也可以设为别的。这一步你明白代码在干啥就行了。训练测试用的类别数是15。
一般自己的数据集都要改输入输出什么的。全卷积虽然可以输入任意长宽,但我们要组成batch,所以长宽是固定的。

代码仅是参考,很多不完善的地方,感觉不合理或有疑惑就改改看。🤣

from keras-segmentation.

Niccccci avatar Niccccci commented on June 9, 2024

十分感谢您的解答,我现在已经完成unet的训练了,之前无法训练是因为我的label图是彩色的8索引图,后来和您的数据集中的label统一格式之后就可以训练了。但是现在遇到了一个问题,我的train的loss先下降再趋于平稳,train的acc也一直升高再趋于平稳,但是val的loss和train一直上下波动,我用训练了300epoch的unet model去做predict,在test数据集上表现不是那么好,但是在train数据集上就分割的很好,这是遇到了过拟合的情况吧?我的数据集是320*320的,train中一共1500张左右,test中300多张,我觉得这个数据数量已经比较多了,请问您知道接下来我该做什么来改变这种过拟合的情况呢?
以下是我训练的曲线图。
image
image

from keras-segmentation.

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.