maxi1571/pharmaceutical-sales-prediction-across-multiple-stores

Pharmaceutical-Sales-prediction-across-multiple-stores

The main objective is to build an end-to-end product that forecast sales in all Rossmann Pharmaceuticals stores across several cities.

Data

• Data is found in https://www.kaggle.com/c/rossmann-store-sales/data
• The data used contains 3 different informations which are the store information and the sales information for each store throughout the data collected years and additional test csv file to test the output.

Columns

• Sales: the turnover for any given day (target variable).
• Customers: the number of customers on a given day.
• Open: an indicator for whether the store was open: 0 = closed, 1 = open.
• Promo: indicates whether a store is running a promo on that day.
• StateHoliday: indicates a state holiday. Normally all stores, with few exceptions, are closed on state holidays.
• SchoolHoliday: indicates if the (Store, Date) was affected by the closure of public schools.
• Store: a unique Id for each store
• 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
• Promo2: Promo2 is a continuing a 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.

Models

1. facebook prophet model used for forcasting
2. Regression model
3. Random Forest
4. Xgboost
5. Deep learning using LSTM

Models output

Linear Regression

Hyperparameters values tuned

	fit_intercept=True,
	normalize=True,
	n_jobs=None
	Metrics values results

Metrics

	Mean squared error on validation data = 0.118
	Mean absolute error on validation data = 0.253
	Mean R2 score on validation data = 0.989

Random Forest

Hyperparameters values tuned

	n_estimators = 12
	Metrics values results

Metrics

	Mean squared error on validation data = 0.0184
	Mean absolute error on validation data = 0.0765
	Mean R2 score on validation data = 0.998

Xgboost

Hyperparameters values tuned

	alpha = 11
	learning_rate = 0.09
	random_state = 4

Metrics

	Mean absolute error on validation data = 0.1839
	Mean squared error on validation data = 0.0687
	Mean R2 score on validation data = 0.993

Deep Learning LSTM

Hyperparameters values tuned

	LSTM = 10
	DNN layer = 2
	Epoch = 2

Metrics

	Mean squared error on validation data = 0.0320
	Mean squared error on training data = 0.0380	

Web App

• Streamlit is used to build the web app.
• A prediction csv file of the test.csv data set is saved to displace as an output on the webapp.

Conclusion

• The quantity of clients is significantly connected with sales.
• During school holidays, more stores are open than during state holidays.
• Promotions result in an increase in both sales and customers for all stores.
• Stores that are open during the school holidays generate more sales than on other days.
• A is the most popular storetype.
• The LSTM Model can do give a better prediction prefromance if layers are increased and training Epochs are increase, but for this expriement the training epoch and hiddlen layers are less.
• The best Model from the Metrics out put is Random forest.