Giter Site home page Giter Site logo

eshanjairath / image-classifier Goto Github PK

View Code? Open in Web Editor NEW
0.0 1.0 0.0 5.72 MB

Image classifiaction done on cifar 10 using deep learning (CNN)

Jupyter Notebook 100.00%
deep-learning convolutional-neural-networks cifar-10 dataaugmentation batchnormalization dropout

image-classifier's Introduction

Image-Classifier

Image classifiaction done on cifar 10 using deep learning (CNN)

In this Repository there are 3 CNN models that are trained on Cifar-10 dataset and all three of them gives diffrerent accuracy in increasing order. The models were tested to predict one image of every category in the CIFAR-10 all models in this study properly predicted the images , although the class probabilities were different. The random images of every category that were used to test the prediction of every model in this study are in the sample images Folder.

In the Case of Model 1 - The accuracy is 78% on validation data using Adam optimizer

In the Case of Model 2 - The accuracy is 82% on validation data using SGD

In the Case of Model 3 - The accuracy is 86% on validation data using SGD

It is clearly visible that changing the optimizer in the case of model 1 affected the accuracy as adam optimizer was able to get higher accuracy. Introduction of data augmentation in the case of model 2 and using test set directly for the validation resulted in 2% increase in accuracy on SGD optimizer which outperformed Adam and resulted in 4% boost in accuracy from the initial model 1. When Batch normalization layers were added after every convolution layers in the case of model 3, the model showed huge 8% jump in validation accuracy, when compared to initial model. Models 2 and 3 predicted all the images of every category correctly with possibilities in increasing order as per model. The initial model with lowest accuracy made one incorrect prediction and gave low possibilities of every category when compared to model 2 and 3.

image-classifier's People

Contributors

eshanjairath avatar

Watchers

 avatar

Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

    ๐Ÿ–– 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. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

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 โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.