Skip to content

victorvsp/kagglePizza

Repository files navigation

Authors: Victor Sanches Portella Allen Paramesh

HOW TO USE

To use the code, first make sure to download the training and test data from https://www.kaggle.com/c/random-acts-of-pizza with the names "test.json" and "train.json", and make sure they are in the same folder as the scripts.

To run the exploratory analysin, use the bash command:

$ python exploratory.py

To run the model, run the command:

$ python script.py

The last one will print the cross-validation AUC score and generate a prediction of probabilities for the test file and will write these prediction in the form "req_id prediction" in the file "testpredict.csv".

To change which model is being used go to the line 174 and choose which model to use.

When using Random Forests to print the OOB score uncomment the lines 264 and 265

About

Code for "Random Acts of Pizza" Kaggle competition. Done as final project for ML course at UNSW.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages