DROP is a system that efficiently performs dimensionality reduction via principal component analysis (PCA) using end-to-end workload optimization. You can refer to our paper for information on how DROP operates.
We have provided a DROP demo script (scripts/DROPDemo.py) and sample dataset from the UCR Time Series Repository (data/labeled/wafer.csv).
From the project root directory, run ./scripts/DROPDemo.py to build and run the demo. The transformed dataset will be written to data/transformed.