Rossmann Pharmaceuticals finance team wants to forecast sales in all their stores across several cities six weeks ahead of time. Managers in individual stores rely on their years of experience as well as their personal judgment to forecast sales.
The data team identified factors such as promotions, competition, school and state holidays, seasonality, and locality as necessary for predicting the sales across the various stores.
The job is to build and serve an end-to-end product that delivers this prediction to analysts in the finance team.
- Exploration of customer purchasing behavior
- Logging
- Prediction of store sales
- Preprocessing
- Building models with sklearn pipelines
- Choose a loss function
- Post Prediction analysis
- Serialize models
- Building model with deep learning
- Using MLFlow to serve the prediction
- Serving predictions on a web interface
- Deployment
The data for this challenge can be found here.
- data: Contains the timeseries sales data
- scripts: Contains script codes
- notebooks: Contains Jupyter notebooks
- models: Contains the trained models
- tests: Unit test files
- ui: Streamlit app
- database: data persistence using mysql
- .github/workflows- Contains yml configration file of github acyion
- .dvc: Data versioning related configration and files
- .travis.yml: Contains config file for travis ci/cd
- Haylemicheal Berihun
- 10academy team