https://github.com/rizwanmunawar/cats-vs-dogs-classification-computer-vision-

Cats vs dogs classification using deep learning. Data augmentation and convolutional neural networks.
https://github.com/rizwanmunawar/cats-vs-dogs-classification-computer-vision-

cats-vs-dogs-classification convolutional-neural-networks data-augmentation deeplearning dropout google-colab-notebook google-colaboratory

Last synced: 3 months ago
JSON representation

Cats vs dogs classification using deep learning. Data augmentation and convolutional neural networks.

• Host: GitHub
• URL: https://github.com/rizwanmunawar/cats-vs-dogs-classification-computer-vision-
• Owner: RizwanMunawar
• Created: 2020-10-01T08:08:22.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2020-11-25T08:51:09.000Z (almost 5 years ago)
• Last Synced: 2024-10-28T08:25:02.405Z (about 1 year ago)
• Topics: cats-vs-dogs-classification, convolutional-neural-networks, data-augmentation, deeplearning, dropout, google-colab-notebook, google-colaboratory
• Language: Jupyter Notebook
• Homepage:
• Size: 1.57 MB
• Stars: 3
• Watchers: 1
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
## CATS vs DOGS Classification using Convolutional Neural Networks and Data Augmentation

### Dataset Details

  you can download dataset from google apis.


#### Dataset Description

Dataset contain 3000 images of Cats and Dogs,

we will train our model on 1700 images,710 images for validation and 604 images for testing.



Training Images of cats = 850


Training Images of dogs = 850


Validation Images of Cats = 352


Validation Images of Dogs = 358


Testing Images of Cats = 304


Testing Images of Dogs = 300


### Overfiting and Underfitting aviodence Techniques Used

1-Data Augmentation (zoom,horizontal_flip,rotation)


2-Dropout


### Model Summary

I used convolutional neural networks with 32, 64 and 128 layers.






Training and Validation Graph:





### Results

Achieved 84% Accuracy on Training data with epochs = 100


81% accuracy on validation data


80% accuracy on testing data