Digits Manipulating of Shiny Project

author: Autoz date: 2018-7-22 autosize: true

Introduction

Coursera-Course-Project-Shiny-Application-and-Reproducible-Pitch

• This Repo contains R codes and Markdown files for Shiny Project of Coursera Developing Data Products course.

• The application contain three part of malipulating and tuning MNIST Digits image data.

• The images were refered to their original IDX format from http://yann.lecun.com/exdb/mnist/. Dataset is from mldata. It is a public repository for machine learning data, supported by the PASCAL network.Fetch MNIST DATA from but mldata seens can not access so redirect to another links.

• Project Shiny App on ShinyApp.io please visit PROJECT LINK.

Outline

• Statistics of the distance of two Digits of certain n umber.
• Comparing with PCA compressed and PCA high dimentions compressed.
• Show the most similar images of Digits with certain Digit.

Shiny App R Codes and Guide on github please visit GITHUB REPO.

Shiny files

• app.R
  • show_compared_img func
  • show_compress_img func
  • show_recommend_img func
  • ui func
  • server func
  • others func
• ShinyDigitSliders.Rpres

Display Compared Images

Display screenshots about Compared Images and Plot summaries distances between two certain Digit numbers.

Display PCA Compress Images

Display screenshots about Compared Images or summary distance between two certain Digit numbers. You can choose PCA (if n > p)and PCA high dimension (if p >> n) method to compress.

Display Similar Images

Display screenshots about Similar Images with certain Digit numbers. It can select different certain Digit numbers and sample range for illustrating how similar with current Digit.

END