Skip to content

cindehaa/autonomy-bootcamp

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 

Repository files navigation


CNN Image Classification

Created for the WARG Autonomy Bootcamp.

This project implements a Convolutional Neural Network (CNN) for image classification. The dataset used was the CIFAR-10 dataset, which consists of 60,000 32x32 color images in 10 classes, with 6,000 images per class. The test set's accuracy was 7074/10000 (71%) after 30 epochs.

ModelTraining Process

Model

The model architectured used is a simple CNN architecture with the following layers:

  • Convolutional layer: 16 filters, kernel size 3, padding 1
  • Batch normalization layer
  • ReLU activation layer
  • Convolutional layer with 32 filters, kernel size 3, padding 1
  • Batch normalization layer
  • ReLU activation layer
  • Max pooling layer: kernel size 2x2, stride 2
  • Fully connected layer: input size 16x16x16, output size 10 (number of classes)

Training Process

  • Data preprocessing: The images are normalized and transformed using data augmentation techniques such as random cropping and horizontal flipping
  • Optimization: Stochastic Gradient Descent (SGD) optimizer is used with a learning rate of 0.005 and momentum of 0.9 to minimize the Cross Entropy Loss and accelerate learning

GitHub @cindehaa  ·  LinkedIn @cindehaa

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%