Skip to content

An end-to-end data science project on Retail Industry. It uses multiple tools available on IBM Watson Studio to do data preparation, data visualization, modeling, model deployment and using the deployed models into front end application.

Notifications You must be signed in to change notification settings

Abeer-Haroon/Predicting-monthly-demand-of-items

Repository files navigation

Predicting monthly demand of items

Prerequisites

  1. Sign up for an IBM Cloud account.
  2. Download or Clone this repo.

Estimated time

This tutorial takes about 20 minutes to complete if you already have an IBM cloud account set up.

Steps

  1. Create an instance of the Watson Machine Learning

  2. Create an instance of the Cloud Object Storage

  3. Create an instance of the IBM Cognos Dashboard Embedded

  4. Create an instance of the Watson Studio

    • Go to the Watson Studio page in the IBM Cloud Catalog.
    • Click Create.

  • Click Get Started.

  • Click Create a Project > Standard

  • Name the project. Select Storage > cloud object storage. Create

  1. Upload the Dataset

  1. Open the dataset. Click on Refine to cleanse and shape it.

    • Click on operation on the left - Select the Column "Item". Select Replace Substring and replace the strings with the following numbers as shown in the below gif.

    • TShirt - 1, Formal Shirts - 2, Jeans - 3, Formal Trousers - 4, Blazers - 5, Jackets - 6, Shoes - 7, Heels - 8, Scarves - 9, Hats - 10

- Select the right data type for every column by clicking on the 3 dots on the column > Convert column type > Select the right data type for the column.

- Save and Run the flow.

- Create an SPSS Modeler Flow. Upload the existing 'SPSS Flow - Store1Forecast.str' from the downloaded folder.

  • Create another flow the same way and upload 'SPSS Flow - Store2Forecast.str'
  • Run the flow

  • Save the created models by clicking on the golden nugget for each model and saving it as a PMML script.

  • Click on the saved model to deploy it as shown below. Repeat this for all the models.

  • Launch RStudio

  • Upload the folder Forecast to RStudio

  • Update Watson Machine Learning credentials and model endpoints in the code and run the app!

About

An end-to-end data science project on Retail Industry. It uses multiple tools available on IBM Watson Studio to do data preparation, data visualization, modeling, model deployment and using the deployed models into front end application.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published