Story and Scenario

You are a data analyst and need to analyze structure(CSV/XLSX/Excel) data. However, you do not know which model is the best and need model management tools to store model information and deploy model service. Then, you can follow this document to train and deploy the model by the open-source low-code tool - PyCaret.

Demo Video

• All functions review demo: Train model, job submit, registry model, deploy model, predict model, and demo in Streamlit:

Pycaret end-to-end demo.mp4

Pycaret Introduction

• Official website:

  Home - PyCaret

• Introduction:

  PyCaret is an open-source, low-code, machine learning library in Python that allows you to go from preparing your data to deploying your model within minutes in your choice of notebook environment.

Dataset - Bank Marketing Dataset

• Kaggle link:

  Bank Marketing Dataset

• Introduction:

  Find the best strategies to improve for the next marketing campaign. We have to analyze the last marketing campaign the bank performed, and identify the patterns that will help us find conclusions to develop future strategies.

Code

• Github showcase:

  showcase/pycaret_classification at feature/sc-21291/pycaret-classification-showcase · InfuseAI/showcase

Part 1: How to use PyCaret to training model on PrimeHub.

Step 1: Create the MLFlow app and setting MLFlow in PrimeHub.

• Document:

  Create an MLflow server · PrimeHub

• Application Setting:

  • Table of setting.

Step 2: Create a Jupyter notebook

Start Notebook · PrimeHub