mhttln/image_classification-vgg-lenet

image_classification-vgg-lenet

Introduction

This work is based on Python 3 and Keras library with TensorFlow backend. Data used in this work is cifar-10 dataset. lenet.py Run the lenet_cifar10.py and smallvggnet_cifar10.py to training and to see the results.

Data

The dataset used is CIFAR-10, which consists of 60,000 32x32 RGB images in 10 classes: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, truck. https://www.cs.toronto.edu/~kriz/cifar.html

Reuslt

Lenet accuracy:63% Smallvgg accuracy:83%

smallvgg result

lenet result