Model to predict wildfires severity class - tbr1, nt13, zdm3
Steps to reproduce result ---
Python3 Modules used - Sklearn, Numpy
-
Clone Repo
-
Download the dataset - https://www.kaggle.com/rtatman/188-million-us-wildfires, place in the cloned repo directory
-
Run firestocsv.py
-
Run code.py
-
Run model.py
The dataset is seperated into groups based off the statistical cause. For each group a Random Forest Classifier is used to attempt to predict the severity class of the fire. Resulting accuracy is printed at the end of each group.