Giter Club home page Giter Club logo

image-classification's Introduction

image-classification

《人工智能基础》第二次大作业,用深度学习网络训练图像分类模型

任务描述

1.本次作业的数据包含30000张图片组成的训练集(train.npy和train.csv)以及5000张图片组成的测试集(test.npy),均可从kaggle网站Data栏目下下载。

2.npy文件可通过numpy.load()函数读取,每个npy文件包含一个N*784的矩阵,N为图片数量。矩阵每行对应一张28*28的图片,同学可在预处理环节自行将每行784维的向量转换成28*28的图像。

3.train.csv文件包含训练集的标签,含image_id和label两列,共30000行,image_id对应矩阵中的行下标,label为该图片的类别标签。

4.在预测环节,需要利用训练好的模型对测试集中的5000张图片进行分类,预测结果应生成submit.csv文件,同样包含image_id和label两列,共5000行,每行对应一张图片的结果。在kaggle网站上提交该文件后会看到自己的分数(指标为catergorization accuracy)及排名,提交文件格式可参考kaggle网站的Data栏目下的samplesummission.csv。

实现细节

1.网络模型选取resnet18,解决梯度消失问题

2.采用GPU加速

3.其他细节见大作业项目报告

image-classification's People

Contributors

jiangz17thu avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar hongyi xie avatar  avatar  avatar

Watchers

 avatar

Forkers

lwwwc hzhou002

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.