This repository contains the data and codes for MSBD_5001_kaggle_competition. All the codes were written based on python programming language.
The train.csv and test.csv are downloaded from the kaggle competition, and the weather.csv is crawled from http://www.worldweatheronline.com/.
---------- data
-------------------- train.csv (containing the train dataset)
-------------------- test.csv (containing the test dataset)
-------------------- weather.csv (containing the weather data)
---------- Xgboost_model.ipynb (containing the xgboost model for preticting the speed of the test dataset)
---------- SVT_model.ipynb (containing the SVT model for preticting the speed of the test dataset)
---------- xgb_speed_prediction.csv (predicted result by using the Xgboost model)
---------- svt_speed_prediction.csv (predicted result by using the SVT model)
---------- ensemble_speed_prediction.csv (ensemble result of Xgboost model and SVT model)
* open the Xgboost_model.ipynb
* the required packages include: numpy, pandas, datetime, warnings, matplotlib, seaborn, sklearn, xgboost
* run all the codes, the submission csv "xgb_speed_prediction.csv" will be saved in the current working path
* and the ensembled result "ensemble_speed_prediction.csv" which is the average of Xgboost model and SVT model could also be got.
* open the SVT_model.ipynb
* the required packages include: numpy, pandas, time
* run all the codes, the submission csv "svt_speed_prediction.csv" will be saved in the current working path
The final submission result is "ensemble_speed_prediction.csv".
Please refer to MIT License Copyright (c) 2020 YJiangcm