- To run the code, please download the dataset from here.
- After download the dataset, please put those files in the folder "data" in the root, and change the sample_submission.csv to submission.csv. You can also rewrite the file path in the code. This part of code always put in the beginning of the file.
- Those codes needs many dependencies, such as pandas, pickle, nltk, numpy, XGBRegressor, LGBMRegressor, sklearn
- Some code may takes time to run, especially those code figuring out the sales tendency. You could run it in terminal directly or use better devices to accelerate.