~                        Ham/Spam Naive Bayesian Classifier 
~						   created by: Robert Herrera
~

1. ---

NOTE: The necessary files to evaluate the necessary evalution1 and evaluation2 idx files have been pre computed and stored as evaluation1_output.txt and evaluation2_output.txt .If you wish to re-evaluate from scratch, please refer to evalution(4)

2. ---

The default dictionary output of the spam filter program is in dictionary.txt. This file is necessary when using the evaluation program.

3. ---

expample script to invoke training program: python train.py --corpus ./trec05p-1/ --key_file ./CompanionFiles2/train.key --train_file ./CompanionFiles2/train.idx

4. ---

example script to invoke evaluation program: python evaluation.py --file ./CompanionFiles2/test1.idx --corpus ./trec05p-1/ -d dictionary.txt --output output.txt

• for example if you were to perform an evaluatin on evaluation.idx, invoke the following: python evaluation.py --file ./CompanionFiles2/evaluation1.idx --corpus ./trec05p-1/ -d dictionary.txt --output eval_output.txt

5. ---

example script to invoke compute performance program: python compute_performance.py --test output.key --target ./CompanionFiles2/test1.key

• for exmpale is you were to perform a test performance on a secret, unknown evaluation.key you would type the following:

  python compute_performance.py --test output.key --target ./Absolute/Path/To/Secret/Key/evaluation.key

   								or
  

  python compute_performance.py --test output.key --target /Users/PDeleon/Absolute/Path/To/Secret/Key/evaluation.key