Rossman Store Sales Prediction

This project was a mini competition done at the Data Science Retreat Berlin.

The goal of this project was to predict the sales number of Rossman stores over time. The data was adapted from the Kaggle Rossman challenge.

The data given includes the daily sales number of Rossman stores from January 2014 to July 2015.

In addition, information on the individual stores is provided, such as the type of store, the assortment of items sold, etc.

Team Members

Maryam Faramarzi
Naveen Korra
Frank Schlosser

Overview

A explanation our methodology and results is given in the notebook notebooks/0_summary.

We performed data exploration, data cleaning and feature engineering of the original Rossman store data.

We trained several models, including Linear Regression, Random Forest Regressors and Gradient Boosted Trees, and performed hyperparameter optimization.

The steps are also described in the notebooks:

Result

We used the root mean square percentage error (RMSPE) as an evaluation metric.

Model	RMSPE
Naive model (mean)	62.03
Linear Regression	22.81
Random Forest	17.07
Gradient Bossted Tree	12.38

The best performing model was a Gradient Boosted Tree implemented with the LightGBM library.

On the holdout dataset of the competition, the GBT achieved an RMSPE of 16.17%.

Comparison of prediction vs. actual sales per day:

Usage

Here we describe how you can recreate the results, using our pre-trained models or by training them yourself.

Setup

First, clone and repository and enter it

git clone git@github.com:franksh/dsr-rossman-comp.git
cd dsr-rossman-comp

Create a virtual environment using Python 3.8, for example using conda.

conda create -n rossman-comp python=3.8.12 pip ipykernel
conda activate rossman-comp

Then install the required packages using

pip install -r requirements.txt

Make the Kernel visible using

python -m ipykernel install --user --name rossman-comp --display-name "rossman-comp"

Using a pre-trained model

We implemented some utility to make it easy to save and load model.

Pre-trained models are stored under data/trained_models/

To evaluate a model, run:

cd scripts
python score_on_holdout.py --holdout_path=path

The holdout_path should point to a file containing holdout data in the correct format.

By default, the best trained model is used. You can specify a different model using the parameter pipeline.

Creating a Kaggle Submission

To create a model prediction and Kaggle submission file, run:

cd scripts
python create_submission.py --pipeline=pipeline_name

You can specify which pipeline to use for the submission.

If you pass best or no argument, the best current model is used.

Different trained pipelines (including the model) are stored under /data/trained_pipelines/.

Additional Information

Dataset

The dataset is consists of two csvs:

#  store.csv
['Store', 'StoreType', 'Assortment', 'CompetitionDistance', 'CompetitionOpenSinceMonth', 'CompetitionOpenSinceYear', 'Promo2', 'Promo2SinceWeek', 'Promo2SinceYear', 'PromoInterval']

#  train.csv
['Date', 'Store', 'DayOfWeek', 'Sales', 'Customers', 'Open', 'Promo','StateHoliday', 'SchoolHoliday']

Description of included columns:

Id - an Id that represents a (Store, Date) duple within the test set

Store - a unique Id for each store

Sales - the turnover for any given day (this is what you are predicting)

Customers - the number of customers on a given day

Open - an indicator for whether the store was open: 0 = closed, 1 = open

StateHoliday - indicates a state holiday. Normally all stores, with few exceptions, are closed on state holidays. Note that all schools are closed on public holidays and weekends. a = public holiday, b = Easter holiday, c = Christmas, 0 = None

SchoolHoliday - indicates if the (Store, Date) was affected by the closure of public schools

StoreType - differentiates between 4 different store models: a, b, c, d

Assortment - describes an assortment level: a = basic, b = extra, c = extended

CompetitionDistance - distance in meters to the nearest competitor store

CompetitionOpenSince[Month/Year] - gives the approximate year and month of the time the nearest competitor was opened

Promo - indicates whether a store is running a promo on that day

Promo2 - Promo2 is a continuing and consecutive promotion for some stores: 0 = store is not participating, 1 = store is participating

Promo2Since[Year/Week] - describes the year and calendar week when the store started participating in Promo2
PromoInterval - describes the consecutive intervals Promo2 is started, naming the months the promotion is started anew. E.g. "Feb,May,Aug,Nov" means each round starts in February, May, August, November of any given year for that store

The holdout test period is from 2014-08-01 to 2015-07-31 - the holdout test dataset is the same format as train.csv, as is called holdout.csv.

MaryamFaramarzi/Rossmann-minicomp