This file explains how to run the WNV_Cardal.py code, which generates the winning model in the Kaggle WNV competition.
System and Dependencies
The code is written in python. I used Python 2.7.7 on Windows 7 Professional. The following python libraries are used by the script: numpy (1.8.1), scipy (0.14.0), pandas (0.14.0), sklearn (0.15.2)
Files
Copy the WNV_Cardal.py script and the SETTINGS.json files to a directory on your system. This directory should also contain the training and test data files and the sample submission file.
Settings
The SETTINGS.json file contains the following parameters:
-
"INPUT_DIR": "." - name of folder with the train.csv, test.csv and sampleSubmission.csv files, as downloaded from Kaggle
-
"SUBMISSION_DIR": "." - name of folder in which the submission file with the results will be saved
Execution
The code should be run from within the directory that contains the script (i.e., you should "cd" into the directory that contains "WNV_Cardal.py" and the input files). Simply run "python Python_Cardal.py" - this will read the input files, create several features, compute prediction probabilities and write the results to a submission file called "wnv_cardal_final.csv". The execution may take several hours, due to the computation of "nearby multirow counts".