haocai1992/MLOps-Specialization

Course notes, quizzes, and programming assignments for DeepLearning.AI's "Machine Learning Engineering for Production (MLOps) Specialization"

MLOps-Specialization

This repository contains all course notes, quizzes, and programming assignments for Coursera MOOC Machine Learning Engineering for Production (MLOps) Specialization, provided by DeepLearning.AI.

About this specialization

Understanding machine learning and deep learning concepts is essential, but if you’re looking to build an effective AI career, you need production engineering capabilities as well.

Effectively deploying machine learning models requires competencies more commonly found in technical fields such as software engineering and DevOps. Machine learning engineering for production combines the foundational concepts of machine learning with the functional expertise of modern software development and engineering roles.

The Machine Learning Engineering for Production (MLOps) Specialization covers how to conceptualize, build, and maintain integrated systems that continuously operate in production. In striking contrast with standard machine learning modeling, production systems need to handle relentless evolving data. Moreover, the production system must run non-stop at the minimum cost while producing the maximum performance. In this Specialization, you will learn how to use well-established tools and methodologies for doing all of this effectively and efficiently.

In this Specialization, you will become familiar with the capabilities, challenges, and consequences of machine learning engineering in production. By the end, you will be ready to employ your new production-ready skills to participate in the development of leading-edge AI technology to solve real-world problems.

Applied Learning Project

By the end, you'll be ready to

• Design an ML production system end-to-end: project scoping, data needs, modeling strategies, and deployment requirements

• Establish a model baseline, address concept drift, and prototype how to develop, deploy, and continuously improve a productionized ML application

• Build data pipelines by gathering, cleaning, and validating datasets

• Implement feature engineering, transformation, and selection with TensorFlow Extended

• Establish data lifecycle by leveraging data lineage and provenance metadata tools and follow data evolution with enterprise data schemas

• Apply techniques to manage modeling resources and best serve offline/online inference requests

• Use analytics to address model fairness, explainability issues, and mitigate bottlenecks

• Deliver deployment pipelines for model serving that require different infrastructures

• Apply best practices and progressive delivery techniques to maintain a continuously operating production system

Timeline

• Date Started: May 31, 2021

• Date Completed: Oct 11, 2021

Certificate

Notes, Quizzes and Lab assignments

Course 1: Introduction to Machine Learning in Production

Week 1: Overview of the ML Lifecycle and Deployment

• Note:
  • Course-Note-C1_W1
• Quiz:
  • The Machine Learning Project Lifecycle
  • Deployment

Week 2: Select and Train a Model

• Note:
  • Course-Note-C1_W2
• Quiz:
  • Selecting and Training a Model
  • Modeling challenges
• Lab:
  • A journey through Data

Week 3: Data Definition and Baseline

• Note:
  • Course-Note-C1_W3
• Quiz:
  • Data Stage of the ML Production Lifecycle
  • Scoping (optional)
• Lab:
  • Data Labeling, PDF

---

Course 2: Machine Learning Data Lifecycle in Production

Week 1: Collecting, Labeling and Validating Data

• Note:
  • Course-Note-C2_W1
• Quiz:
  • Intro to MLEP
  • Data Collection
  • Data Labeling
  • Issues in Training Data
• Lab:
  • TFDV Exercise, PDF
• Assignment:
  • Data Validation, PDF

Week 2: Feature Engineering, Transformation and Selection

• Note:
  • Course-Note-C2_W2
• Quiz:
  • Feature Engineering and Preprocessing
  • Feature Transformation
  • Feature Selection
• Lab:
  • Simple Feature Engineering, PDF
  • Feature Engineering Pipeline, PDF
  • Feature Selection, PDF
• Assignment:
  • Feature Engineering, HTML

Week 3: Data Journey and Data Storage

• Note:
  • Course-Note-C2_W3
• Quiz:
  • Data Journey
  • Schema Environments
  • Enterprise Data Storage
• Lab:
  • ML Metadata, PDF
  • Iterative Schema, PDF
• Assignment:
  • Data Pipeline Components for Production ML, HTML

Week 4 (Optional): Advanced Labeling, Augmentation and Data Preprocessing

• Note:
  • Course-Note-C2_W4
• Quiz:
  • Advanced Labelling
  • Data Augmentation
  • Different Data Types
• Lab:
  • Feature Engineering with Weather Data
  • Feature Engineering with Accelerometer Data
  • Feature Engineering with Images, PDF

---

Course 3: Machine Learning Modeling Pipelines in Production

Week 1: Neural Architecture Search

• Note:
  • Course-Note-C3_W1
• Quiz:
  • Hyperparameter Tuning and Neural Architecture Search
  • AutoML
• Lab:
  • Intro to Keras Tuner
  • Hyperparameter tuning and model training with TFX

Week 2: Model Resource Management Techniques

• Note:
  • Course-Note-C3_W2
• Quiz:
  • Dimensionality Reduction
  • Quantization and Pruning
• Lab:
  • Manual Feature Engineering
  • Algorithmic Dimensionality Reduction
  • Quantization and Pruning

Week 3: High-Performance Modeling

• Note:
  • Course-Note-C3_W3
• Quiz:
  • High-Performance Modeling
  • Knowledge Distillation
• lab:
  • Distributed Strategies with TF and Keras
  • Knowledge Distillation

Week 4: Model Analysis

• Note:
  • Course-Note-C3_W4
• Quiz:
  • Model Analysis
  • Model Analysis and Debugging
  • Continuous Evaluation and Monitoring
• Lab:
  • TensorFlow Model Analysis
  • Model Analysis with TFX Evaluator
  • Fairness Indicators

Week 5: Interpretability

• Note:
  • Course-Note-C3_W5
• Quiz:
  • Explainable AI
  • Interpretability
  • Understanding Model Predictions
• Lab:
  • Shapley Values
  • Permutation Feature Importance

---

Course 4: Deploying Machine Learning Models in Production

Week 1: Model Serving: Introduction

• Note:
  • Course Note-C4_W1
• Quiz:
  • Introduction to Model Serving
  • Introduction to Model Serving Infrastructure
  • TensorFlow Serving
• Lab:
  • Intro to Docker and installation
  • First look at Tensorflow Serving with Docker
  • Serve a model with TensorFlow Serving
  • Build, train, and deploy an XGBoost model on Cloud AI Platform

Week 2: Model Serving: Patterns and Infrastructure

• Note:
  • Course Note-C4_W2
• Quiz:
  • Model serving architecture
  • Scaling Infrastructure
  • Online Inference
  • Data Preprocessing
  • Batch inference scenarios
  • Batch Processing with ETL
• Lab:
  • Deploy a ML model with fastAPI and Docker
  • Practice Kubernetes in your Local Environment
  • Load testing servers with Docker Compose and Locust
  • ETL Pipelines and Batch Predictions with Apache Beam and Tensorflow

Week 3: Model Management and Delivery

• Note:
  • Course Note-C4_W3
• Quiz:
  • ML Experiments Management and Workflow Automation
  • MLOps Methodology
  • Model Management and Deployment Infrastructure
• Lab:
  • Building ML Pipelines with Kubeflow
  • Developing Custom TFX Components
  • Model versioning with Tensorflow Serving and Docker
  • Intro to CI/CD pipelines with GitHub Actions

Week 4: Model Monitoring and Logging

• Note:
  • Course Note-C4_W4
• Quiz:
  • Model Monitoring and Logging
  • Model Decay
  • GDPR and Privacy

Disclaimer

• Copyright of all materials in this repository belongs to DeepLearning.AI, and can only be used or distributed for educational purpose. You may not use or distribute them for commercial purposes.
• My quiz and assignment solutions are for reference only. Please do not copy any part of the code/answer directly.