➡️ REPORT LINK
➡️ GitHub Repository - https://github.com/AngRoy/BrainDead2k24_Problem1-
Concatenate the two tables of the problem 1 dataset - https://www.kaggle.com/datasets/braindeadiiest/brain-dead-rice-production-data-in-india?rvi=1
-
Run the notebook ps1_data_preprocessing.ipynb on this concatenated csv file to obtain processed_ds1.csv and pre_processed_ds1.csv
-
The jupyter notebook Problem1Analysis_EDA.ipynb is used for all the analysis tasks. Run this notebook on processed_ds1.csv file for EDA.
-
For creating Cluster Map of Rice Production we needed data such that it doesnot contain any NaN values neither any empty values, one approach was to replace those cells with the column median but for this analysis we used processed_ds1_telengana_predicted.csv where by analysing the future trends of the state earlier values have been predicted by our model.
- Run the notebook Forecast_without_AP&Telangana.ipynb on processed_ds1.csv to forecast rice production of all states except Andhra Pradesh & Telangana for the next 5 years (2023-28).
- Run the notebook AP_+_Telangana_Rice_Prod_Forecast.ipynb on processed_ds1.csv to forecast rice production of Andhra Pradesh & Telangana for the next 5 years (2023-28). This notebook also has script to obtain plots of forecast.
- Run the notebook External_Dataset/rice.ipynb on External_Dataset/rice_aggregated_2004_2020.csv to analyse how rice production is affected by external factors such as rainfall etc.