19 -- Spamalot -- Karthik Kasala

[kasalak]

• [x] Blank slate
  • [x] Create a GitHub repo called spambase
  • [x] Set up your requirements/virtual environment
  • [x] Download the spambase dataset, but do not commit it to your repository
  • [x] Create an IPython notebook to write your code and collect your findings
• [x] Normal mode
  • [x] Load in the data file, doing any cleaning necessary to get usable data
  • [x] Subsample the data set into training and test data
  • [x] Write code to classify the data into spam/not-spam, making sure that you just use your training data to build your model, and checking your results with your test data
• [ ] Hard mode
  • [ ] Try reducing or changing your features in order to get better results
  • [ ] Find another dataset and break it down into features
  • [ ] Test your algorthm on the new dataset and compare its performance

[kasalak]

https://github.com/kasalak/Spambase

[powder-river]

@kasalak pretty straight forward, not much to comment on, good job!