Modeling shopping time duration for inferece, includes:
- understand the data via exploration (EDA),
- design a workflow to transform raw data into the feature space of the model,
- build model and predict
- extra: inference of resulting feature space and model
- L1 Regularized Regression (Lasso)
- Random Forest (minor exploration)
preprocessing.py - contains Preprocessing class to:
- Preprocessing of data prior to model fit
- Returns feature engineered datasets
- Handles dummy variables without leakages
- Handles scaling
model.py
- Loads data
- Featurizes data
- Runs Lasso Model
- Outputs predictions into folder ./predictions/
- In console, navigate to ./src/ folder.
- run -model.py
- Python
- Pandas
- Matplotlib / Seaborn
- sklearn
- StatsModels
- Jupyter Notebook