spam-classifier

How to Build a Spam Classifier using Keras in Python

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