Develop ML Models with Python and T-SQL
39mAdvanced2025-02-03
Authors
Sam Nasr
Course details
In this course, senior software engineer and instructor Sam Nasr guides you through the skills you need to leverage Python and T-SQL for building robust machine learning models. Discover how to set up your development environment, import data seamlessly into SQL Server, and prepare it for model training. Explore the fundamentals of various machine learning algorithms, data pre-processing techniques, and tuning hyperparameters for optimal performance. Learn how to use tools such as Visual Studio Code, Pandas, SQL Alchemy, and scikit-learn for effective model development and deployment. Plus, find out how to store, access, and manage version history of your trained models directly within your SQL Server database. By the end of the course, you will be equipped to train, evaluate, and deploy machine learning models adeptly, and implement best practices for data handling and model management.
Learning objectives
Identify how to configure SQL Server to use Python code in T-SQL scripts.
Articulate how to ttrain, evaluate, and use a ML model to make predictions using Python and T-SQL.
Understand how to use Python libraries to train a ML model using data in a SQL Server database using Python and T-SQL.
Determine how to store and retrieve a functional ML model in a SQL Server database using Python and T-SQL.
Learning objectives
Identify how to configure SQL Server to use Python code in T-SQL scripts.
Articulate how to ttrain, evaluate, and use a ML model to make predictions using Python and T-SQL.
Understand how to use Python libraries to train a ML model using data in a SQL Server database using Python and T-SQL.
Determine how to store and retrieve a functional ML model in a SQL Server database using Python and T-SQL.
Skills covered
SQL ServerDatabase AdministrationMachine LearningDatabase DevelopmentPythonDatabase ManagementArtificial Intelligence (AI)Programming LanguagesOpen SourceMicrosoftSoftware DevelopmentOne-Off
Concepts
0. Introduction
- 01 - Unlocking potential - Python and SQL Server for ML
- 02 - What you should know
1. Overview
- 03 - Setting up the development environment
2. Project Setup
- 04 - Model development
- 05 - Python libraries and modules
3. Using the Model
- 06 - Project overview
- 07 - Demo - Training the model for linear regression
- 08 - Demo - Using the model for linear regression predictions
- 09 - Demo - Loading classification data
- 10 - Demo - Training and using the model for classification predictions
- 11 - Demo - Loading data for sentiment analysis
- 12 - Demo - Training and using the model for sentiment analysis predictions
- 13 - Demo - Loading housing data for pipeline, cross validation, and hyperparameters
- 14 - Demo - Using data for pipeline, cross validation, and hyperparameters
4. Common Pitfalls and Next Steps
- 15 - Common pitfalls
- 16 - Best practices
- 17 - Next steps