Dog-Classification-with-Tensorflow

Overview

• This is an end-to-end deep learning project using Tensorflow and Keras.
• A dataset of dog images is read, and then a convolutional neural network is trained to classify the dogs by breed.
• This repository contains all the codes and resources used to build this dog classification system.

Steps

• Loading in data with TensorFlow and Keras
• Exploring images in data with PIL (Python Imaging Library)
• Training a neural network with TensorFlow and Keras
• Looking at model error
• Improving model accuracy with dropout
• Augmenting data with Keras
• Investigating model error

Code

You can find the code for this project here.

• dog classification.ipynb

Technologies/Tools

• JupyerLab / Google Colab
• Python 3.10.12
• Tensorflow - pip install tensorflow
• Python packages
  • Pandas - pip install pandas
  • Pillow - pip install pillow
  • Matplotlib - pip install matplotlib

Data

Data used are from the Stanford

You can download the dog image dataset used in this project here.

• dog_images.zip - please unzip this file into a folder called images.

Output