Dog Convolutional Neural Network Classifier

This repository contains details about my project.

Why I made this ?

Suppose there's a dog show going to occur in your city and you're going to volunteer in it. They give the department where you have to register dogs along with their characteristics such as color , size etc. Now there may be some mischevious people with cunning nature who would register dogs that aren't even dogs.So I made this classifier to know which are dogs and which aren't.And that along their breed classification.

Problem statement :

In this Section we are implementing Convolution Neural Network(CNN) Classifier for Classifying dog and cat images. The Total number of images available for training is 25,000 and final testing is done on seperate 10,000 images.

Note:This problem statement and dataset is taken from this Kaggle competition.

Dependencies

• Jupyter notebook
• Tensorflow 1.10
• Python 3.6
• Matplotlib
• Seaborn
• Scikit-Learn
• Pandas
• Numpy

Install dependencies using conda

Test Train Split

Image training set contain 12500 images for each category. I split those into 80% train and 20% means test Split each class images into 10,000 for train and 2,500 for test.

Architecture

#To run this you will need:

1. Dataset of dogs
2. Python knowledge

Thank you for visiting my repository