Two milestones #2
Description
- procedure for analyzing/plotting results out of results (eg Apache0001.csv)
- exhaustive exploration of configuration/hyper-parameters for classification/decision tree
It should be the standard behavior: based on a subject system, it first produces all results; then we can exploit new data.
I would start by first integrating the "analysis" Python scripts (to verify that the data generated are exploitable).
I highly suggest to have a quick and dirty approach: simply integrate what we already have; then we will refactor/modularize/optimize the Docker or scripts.