Skip to content

Fully configurable Deep Convolutional Neural Network for Image Classification, trained by Backpropagation. Implementation does not use of any Deep Learning package (Tensorflow, PyTorch, etc), only Numpy.

Notifications You must be signed in to change notification settings

yBeOne/Image-Classifier-Deep-CNN

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 

Repository files navigation

Image-Classifier-Deep-CNN

Fully configurable Deep Convolutional Neural Network for Image Classification, trained by Backpropagation.

Implementation does not use of any Deep Learning package (Tensorflow, PyTorch, etc), only Numpy, and the number, order, type and parameters of each layer is fully configurable.

Training results of 4-layer CNN (CONV -> ReLu -> Pool -> Fully Conn) trained to classify 10 classes (road signs) given as 52x52 RGB images, 2000 images per class.



Test results on a different set of images, 100 images per class, showing good generalization performance.



About

Fully configurable Deep Convolutional Neural Network for Image Classification, trained by Backpropagation. Implementation does not use of any Deep Learning package (Tensorflow, PyTorch, etc), only Numpy.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%