20 -- Polyglot -- Adam Hartz

[ahartz1]

• [x] Blank slate
  • [x] Create a GitHub repo called programming-language-classifier
  • [x] Set up your requirements/virtual environment
  • [x] Copy all the files from this folder into your repo
• [x] Normal mode
  • [x] Collect a corpus of programming language snippets to use for training data
  • [x] Extract features from your corpus for your classifier, either using an existing Transformer or one you've written
  • [x] Be sure you can explain at a high-level how your Transfomer works
  • [x] Write a classifier function that takes a string of code and returns a guess for the language the code was written in
  • [x] Test your classifier using the provided test files
  • [x] Be sure you do not use the test files as your training data
  • [x] Demonstrate the use/results of your classifier in an IPython notebook
• [ ] Hard mode
  • [ ] Create a runnable Python file that takes a command-line argument of a source code file and prints out the guessed programming language
  • [ ] Calculate the probability of various programming languages, and return the top percentages probable
  • [ ] Add command-line flags to make your program more versatile

[ahartz1]

SUBMIT https://github.com/ahartz1/programming-language-classifier @jamesmallen @powder-river