-------------------------------------------------------------------------------
Paul J. Atzberger
http://atzberger.org/

Olivetti Face Dataset.  Put together for data science course.

Credits: Images from AT&T Laboratories Cambridge and Kaggle.
https://www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html
-------------------------------------------------------------------------------
Data is in the format of python numpy archive files .npy.

- image_data are 400 images of 64x64 pixels. 
- target_labels are 400 labels in range 0, 39 for person shown.

All images can be found in the olivetti_faces.npy. Corresponding 
labels in olivetti_faces_labels.npy.

The data has been split into 20% test dataset and 80% training dataset.  These 
can be found in the files with the corresponding "test" and "train" filenames.
-------------------------------------------------------------------------------
