It is project, where I anlyse influence of forest fires in Amazonia this year(2019). Here you can find whole data processing pipeline, including data mining, integrating, clearing, reducing, handling missing data and outliers. After all data preparation I was using ARIMA model to predict weather for 2019 year and compare prediction with real weather measures.
Here you can find .ipynb files with all data preparation process and analysis. Actually, you don't need to clone this repository to see preparation and analysis, but if you want to run it locally, here are instructions:
Here are some requirements of your system to run code:
- Python3.0+
- Python libraries:
- pandas
- numpy
- matplotlib
- statsmodels
To download this repository on your computer just copy this into your command line:
git clone https://github.com/franchukpetro/Forest-fires-influence-analysis
To run code open folder with clonned repository and run this:
jupyter notebook
Then use browser to run code and see results.
Dataset description: https://data.noaa.gov/dataset/dataset/global-surface-summary-of-the-day-gsod
Full dataset: https://www.ncei.noaa.gov/data/global-summary-of-the-day/archive/
Short version of dataset, used for analysis: https://drive.google.com/open?id=1GHCVPrhszTthhEpzF_NIY-MZHr-NYvDz
Petro Franchuk