Jegge2003/Team-17-Azubi-Africa-Hackathon

Team-17-Azubi-Africa-Hackathon # DOCUMENTATION ON DR CONGO FIRES Main aim of our predictive module is to find the percentage area that will get burnt in case of a fire outbreak. This project is solely about using some specific and environmental indicators to predict the percentage of an area that will get burnt in the case of a fire outbreak. The various steps we used are elaborated more in the Team-17-Report.docx which is the main report. The various processes used in the .ipynb file is commented to give an overview of all that was involved in the process. The main process used in the .ipynb file are: 1. Understanding The Data And Cleaning 2. Feature Engineering And Vizualizations 3. Azure ## Understanding The Data And Cleaning The whole idea of this process is to get an overview of the data and all that it entails. This was done uses various python libraries such as Pandas, Numpy, MissingNo, and many others. The various methods used in this particular broad process includes the following: 1. Imported the need libraries which are; numpy as np, pandas as pd, collections, matplotlib.pyplot as plt, missingno as msno, and scipy.stats as st. 2. Read the data with the read_csv method of pandas which loaded the csv file of our train dataset into a Pandas Dataframe and stored it in the variable train. 3. Showed the first 5 rows by calling the head method on the Pandas Dataframe we named train. 4. Called the describe method on train to give description of the data. 5. Called the info function on train to give an information about the data. 6. Used the isnull().sum() on train to check the total count of null values availabe. 7. The shape method on train was used to get the dimensions of the dataset. 8. The index method on train used to the start and end of the rows of the train dataset 9. The skew function on train to check the skewness of the dataset 10. The kurt function on train to check the kurtosis of the dataset. 11. Run a for loop of all the columns of the dataset an …