Building a Reproducible Model Workflow

Description

In this course, you will learn:

• Dive into reproducible model workflows and machine learning operations while learning about use cases, the history of machine learning, and what you'll construct at the end of the course.
• Create machine learning pipelines and learn how to version data and model artifacts.
• Create reusable processes for exploratory data analysis (EDA), data cleaning and pre-processing, and data segregation/splitting.
• Validate data using deterministic and non-deterministic testing, and experiment with different parameters using PyTest.
• Create an inference pipeline, validate and select the best performing models from experiments, and validate and test your final model artifacts.
• Create an end-to-end pipeline, release it, and deploy it with MLflow.
• Create a reusable end-to-end pipeline for anticipating New York City short-term rental prices!

Syllabus:

• Introduction to Reproducible Model Workflows
• Machine Learning Pipelines
• Data Exploration and Preparation
• Data Validation
• Training, Validation and Experiment Tracking
• Final Pipeline, Release and Deploy
• Build an ML Pipeline for Short-term Rental Prices in NYC