I am using anaconda to run the ipynotebooks locally since they take 24-32 hours to fully run. Will be pushing my work in progress so that others can see where to pick up from
If anyone figures out the errors I am having, please push the changes here.
A suggestion to study different ways to look at modeling and classifying the DDoS dataset if to
- have different CSV files in the train versus test
- Play around with the hyper parameters of Pycaret
you can fork this repo to publish different versions and outputs so we can compare similar to the kaggle interface. An ideal form would be to use sagemaker AWS. but let's try this way to collaborate for local datascience dev for now