Skip to content

ConvNet and PCA/SVM approaches to the kaggle digit recognizer challenge

Notifications You must be signed in to change notification settings

hc2116/Kaggle_Digit_Recognizer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

Kaggle_Digit_Recognizer

ConvNet and PCA/SVM approaches to the kaggle digit recognizer challenge https://www.kaggle.com/c/digit-recognizer/data.

First approach uses a nonlinear SVM (ovo-shape) on the first 100 PCA components, done in Python with SciKitLearn (91% accuracy)

The second approach implements a Convolutional Neural Network with two hidden layers, done in Python with Tensorflow (99% accuracy)

About

ConvNet and PCA/SVM approaches to the kaggle digit recognizer challenge

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages