Created for the Women in Data Science ATX meetup, 9/29/16. This workshop is focused on exploring dimensionality reduction through PCA and SVD. Presented by Randi Ludwig and Julia Galstad.
The associated slide deck is found at: http://docs.google.com/presentation/d/1yUHj71GrG4FHMxnr7OGPNkh84fENkzngdVlxSIixYLs/
icecreamPCA is a subtree from @randirl17 including the following files:
IceCreamSundaes.csv - ~1000 rows of ingredient combinations for ice cream sundaes.
IceCreamData.ipynb - Randomly generate a sample of ice cream sundae combinations. Output is IceCreamSundaes.csv.
icecreamPCA.ipynb - Workbook for exploring the IceCreamSundaes.csv data set, and performing principal component analysis. The overall objectives include deciding how many dimensions to reduce to, how to perform the coordinate transformation and reconstruct the data using the principal components, and how to interpret the principal components to inform data mining and predictive modeling efforts.
PCA_of_an_image_using_SVD is a subtree from @juliathebrave including the following files:
SVD_Blair.ipynb explores an image of Tony Blair using PCA/SVD.
Blair.csv contains a preprocessed image obtained from the dataset used in "Labeled Faces in the Wild". http://vis-www.cs.umass.edu/lfw/lfw-funneled.tgz
figure_1.png is an example of a figure that could be produced using the notebook.