In this project, we'll walk through an end to end deep learning project using Tensorflow and Keras. We'll read in a dataset of dog images, then train a convolutional neural network to classify them by breed.
By the end, you'll know how to use keras to train and optimize a neural network. You'll also learn about how to work with images using Python.
You can find the code for this project here.
File overview:
classifier.ipynb
- a Jupyter notebook that loads the images and trains a neural network.
To follow this project, please install the following locally:
- JupyerLab
- Python 3.8+
- Python packages
- tensorflow
- Pillow
- pandas
- matplotlib
You will also need to have a GPU on your machine and configured. To set things up, you'll need to install GPU support for tensorflow.
If you have issues installing tensorflow and/or don't have a GPU, please use Google Colaboratory. Colaboratory will give you a Jupyter notebook in the cloud with full GPU support.
You'll need to download the dog image dataset to follow this project:
- dog_images.zip - please unzip this file into a folder called
images
.
The data is originally from Stanford. The original dataset has many more breeds included, which you can use to extend your analysis.