Walmart Store Sales forecasting
- Predict Walmart sales for each store and department in it and to predict which stores are affected and the extent of the impact due to holidays or markdowns (Promotional Events)
- Predict weekly sales of the department for store
- Make a Rest API
Assumptions - markdowns precede prominent holidays, the four largest of which are the Super Bowl, Labor Day, Thanksgiving, and Christmas
Output :
- This analysis has been done using neural network.
- Analytical tools used in project are python, scikit learn and keras.
- User interface is developed by Flask web application
Python 3
Main Libraries Used -
- Tensorflow
- Scikit-learn
- Numpy
- Flask
- Pandas
- Keras
Folders Used -
- flask - For Flask Application
- flask/model - trained and exported model
- result - prediction result
- data - having data in csv
$ cd walmart_sales_prediction
$ pip install -r requirements.txt
For Running Flask Application
```sh
$ python flask/api.py
Use http://0.0.0.0:5000/ for web application
-
Data is from Kaggle
-
Using machine learning model, a sales model is made which will help in prediction of sales
-
Following features or parameters are used โ Store, Department, Date, Holiday, Temperature, Fuel Price, Markdown Events, CPI, Unemployment
-
Pre-processing and normalization of data is done
-
Keras model is trained on normalized data
-
Validation accuracy is calculated and model is exported with test data sales output
Constraints / Notes ::
-
Website made where we can submit store, department and date value and if it is special holiday or not and the outcome will be weekly sales value
-
Web application is a user interface which can be easily use by managers for forecasting sales of given date