This project forms part of the coursework for Udacity's Data Science Nanodegree. Read on for more information on how this project is structured, or check out my blog post which summarises the best bits: https://m 69BE edium.com/@aell039/what-dog-breed-do-you-look-like-bda24a56543a
The motivation for this project is to create a convolutional neural network with the ability to classify dogs by their breed, and also to make fun predictions using images of humans.
- Keras - to create our neural network
- OpenCV (cv2) - image manipulation and associated functions
- matplotlib - for drawing bounding boxes on images
- PIL - loading images
haarcascades Face detector model, used to determine if we are working with an image of a human images Imagery used in the notebook (not training or testing images) requirements yml files for system setup saved_models checkpoints of the best models test_images A selection of images to test the model with LICENSE.txt Licensing info README.md What you're reading now dog_app.ipynb Jupyter notebook containing all the steps of this project extract_bottleneck_features.py Helper function to extract bottlenecks
The notebook goes into more detail, but to summarise: the code in this successfully creates and trains a model to determine dog breed with over 80% accuracy. The final cells of the notebook roll this into a function that can be called on any source image you like.