This is a Deep Learning project that aims to classify images as containing either a cat or a dog. The project uses CNN, a type of neural network commonly used for image classification tasks.
CNN are a type of neural network that are designed to automatically learn features and patterns in image data. The architecture of a CNN consists of multiple layers, including convolutional layers, pooling layers, and fully connected layers.
The convolutional layers are the core of the network and apply filters to the input image to extract features. These features are then passed through pooling layers, which reduce the spatial size of the output and help to generalize the features learned by the network. Finally, the output is passed through fully connected layers, which use the extracted features to classify the image.
The main advantage of CNN over other types of neural networks is their ability to effectively extract features from images. The convolutional layers in the network are able to learn filters that detect edges, corners, and other basic features in the image, while deeper layers can learn more complex features such as patterns and textures.
The dataset used for this project is "The Asirra dataset," which consists of over 25,000 images of dogs and cats annotated images of cats and dogs.
The repository contains various files including :
-
Cats_vs_Dogs.ipynb : Jupyter notebook containing the code for the CNN model developed in TensorFlow.
-
cat.jpg : Example image of the input data used to test the models.
-
images_results_cats_dogs.png : Visualization of the results obtained from the notebook.
This project was created by raynaldlao and LH-devcat. The dataset used for this project is "The Asirra dataset," which is available in Kaggle. Here are the URL to retrieve the dataset : https://www.kaggle.com/c/dogs-vs-cats/data