I want to learn more about Machine Learning for images. It always helps to have a project to apply the theory and it's a bit less paint-by-number than online tutorials which helps me learn better.
I am experimenting in jupyter notebooks for now and it's very much a work in progress.
- getting_started.ipynb: exploring the available data, playing with openCV
- data_augmentation.ipynb: functions to split, rotate, and setup the data in the folders expected by Keras/TensorFlow
- binary_classification_pipeline.ipynb: setting up the CNN model, training, validation
- journal.md: notes and thoughts on the project (not organized)