sandy4321 / kaggle_m5_forecasting-1

My first Kaggle competition "m5_forecasting"

m5_forecasting

==============================

Project Caveats

• This is my first ever competition and github repo ... so be critical. I know a lot of mistakes were made so constructive criticism/feedback is much appreciated.
• There are a lot of areas for improvement. The MAIN objective of this was for me to learn.
• Thanks in advance!

Project Overview

• Use hierarchical sales data from Walmart to forecast daily sales for next 28 days.
• More details https://www.kaggle.com/c/m5-forecasting-accuracy/overview/description
• Hierarchical modeling using lgbm at pre-defined strata

Run order:

python ./src/group_level.py 
python ./src/item_level.py
python ./src/final_scale.py

Project Organization

---

├── LICENSE
├── README.md                   <- The top-level README for developers using this project.
├── data
│   ├── processed               <- The final, canonical data sets for modeling.
│   └── raw                     <- The original, immutable data dump.
│
├── models                      <- Trained and serialized models, model predictions, or model summaries
│
├── requirements.txt            <- The requirements file for reproducing the analysis environment, e.g.
│                                  generated with `pip freeze > requirements.txt`
│
├── setup.py                    <- makes project pip installable (pip install -e .) so src can be imported
├── src                         <- Source code for use in this project.
│   ├── __init__.py             <- Makes src a Python module
│   │
│   ├── data                    <- Scripts to download or generate data
│   │   └── etl.py              <- Tranforms raw data, creates lag variables, applies SMOTE for imbalanced data
│   │
│   ├── models                  <- Scripts to train models and then use trained models to make
│   │   │                          predictions
│   │   ├── predict_model.py    <- Make predictions using defined model parameters
│   │   ├── train_model.py      <- Hyperparameter runing using RandomizedSearchCV
|   |   └── lstm_class.py       <- Work in progress. Attempt to incorporate/learn LSTM.
|   |
|   ├── paths.py                <- Generates relative file paths
|   ├── group_level.py          <- Creates forecasts for group/strata (@ state, store, category, and department levels)
|   ├── item_level.py           <- Creates forecasts for all items
|   ├── final_scale.py          <- Hierarchical scaling (state --> state/store --> state/store/category --> 
|   |                              state/store/category/dept --> item
|   ├── compare_models.py       <- Work in progress. Wanted to learn LSTM and see if it produced better results. Will need 
                                   gpu to run this module. Even then, it'll take a significant amount of time.

---

Project structure based on trimmed version of cookiecutter data science project template. #cookiecutterdatascience