Skip to content

Latest commit

 

History

History
 
 

spam-classifier

To run this:

  • pip3 install -r requirements.txt
  • For training, since we're using transfer learning, you first need to download, extract GloVe and put to data folder, this is a pre trained embedding vectors that map each word to its vector, two words that have similar meanings tend to have very close vectors, and so on.
    python3 spam_classifier.py
    
    This will spawn tensorflow logs in logs folder, as well as the model and tokenizer in results, so test.py will use them.
  • After the training has finished, try testing your own emails, or change the code on your needs, or whatever:
    python3 test.py