Cats and Dogs classification

---

In this project we write an algorithm that classify whether an image is cat or dog from the Cats and Dogs Dataset.

Data

Creat a folder called Datasets, download Dataset from https://www.kaggle.com/tongpython/cat-and-dog. Put the data in the Dataset folder and run your code here.

Important Information

---

-Used Python Version: 3.6.0

-install  modules with ```sudo pip3 install -r requirements.txt``` command.

---

Model Architecture

-Input Data Shape : 224*224*3

I implemented a 2 layer convolutional neural network to classify images from the Cats and Dogs Dataset.

I later implemented a 4 layer convolutional neural network on the same Dataset.

Activation function : Rectified Linear Unit (ReLU)

Max Pooling shape : 2 X 2

Classification : log_softmax

Optimizer : Gradient descent with Momentum

---

Projection (visualization) of the model is shown as:

---

Images of Cats and Dogs Classification

---

Loss of the Model

Run FIle : python main.py

---