English | MP4 | AVC 1280×720 | AAC 44KHz 2ch | 35 Lessons (4h 10m) | 606 MB
With both machine learning and DevOps at the forefront these days, Milecia McGregor helps engineers understand how to apply key DevOps principles to their machine learning projects.
When teams are working with machine learning models, changing features, different data sets, new algorithms, and unique computing resources all influence a machine learning model’s performance. Tracking all of these items can be complicated. With tools such as DVC, MLFlow, AWS, you can meet the challenge. Milecia McGregor demonstrates how to use MLOps tools to improve machine learning and automate some of the steps in the process.
What You Will Learn:
Developers and Engineers will learn how to:
- Capitalize on MLOps as an emerging field. Data-focused companies are looking for engineers with these skill sets.
- Build a basic MLOps pipeline from scratch with open-source tools – take a working template with you for your own projects
- Take ChatGPT into account to provide a practical bridge for engineers and DevOps teams.
Table of Contents
Introduction
1 Learn MLOps for Machine Learning Introduction
Lesson 1 Learning the MLOps Pipeline
2 Learning objectives
3 Gather the data
4 Analyze the data
5 Prepare the data
6 Train a model
7 Evaluate the model
8 Validate the model
9 Deploy the model
10 Monitor the model
Lesson 2 Handling the Data
11 Learning objectives
12 Determine what the data sources are
13 Create ETL pipelines to compile the data
14 Understand the data schema with respect to the model
15 Identify data that can be used for the model
16 Perform feature engineering
17 Version the data with DVC
18 Make multiple data sets
19 MLOps best practices for data
Lesson 3 Creating a Model
20 Learning objectives
21 Use common Python libraries
22 Code versioning with Git
23 Perform hyperparameter tuning
24 Track experiments with MLFlow
25 Track experiments with DVC
26 Evaluate the models
Lesson 4 Working with Production Models
27 Learning objectives
28 Decide the best deployment method
29 Test on pre-production environments
30 Deploy to production
31 Monitor the model for drift
32 Validate the pipeline flow
33 Automation points in MLOps
34 Set up redeploy pipeline
Summary
35 Learn MLOps for Machine Learning Summary
Resolve the captcha to access the links!