MLOPS COURSE IN JAMSHEDPUR

MODULE 1: MLOPS INTRODUCTION

• MLOps Overview
• Machine Learning Lifecycle
• Challenges of Tradition Machine Learning lifecycle
• MLOps as a solution.
• MLOps Core Concepts
• MLOps standards and principles

MODULE 2: MLOPS CI/CD/CT PIPELINES

• ML models in Production
• MLOps Continuous Integration (CI)
• MLOps Continuous Delivery (CD)
• MLOps Continuous Training (CT)

MODULE 3: MLOPS MATURITY LEVELS

• Maturity levels, why is it important?
• Various MLOps Maturity Levels
• MLOps Maturity Level 0
• MLOps Maturity Level 1
• MLOps Maturity Level 2

MODULE 4: MLOPS PLATFORMS

• MLOps Architecture
• MLOps Platforms and Tools
• Microsoft Azure ML Foundation
• AWS SageMaker for MLOps

MODULE 1: PYTHON BASICS

• Introduction of python
• Installation of Python and IDE
• Python objects
• Python basic data types
• Number & Booleans, strings
• Arithmetic Operators
• Comparison Operators
• Assignment Operators
• Operator’s precedence and associativity

MODULE 2: PYTHON CONTROL STATEMENTS

• IF Conditional statement
• IF-ELSE
• NESTED IF
• Python Loops basics
• WHILE Statement
• FOR statements
• BREAK and CONTINUE statements

MODULE 3: PYTHON DATA STRUCTURES

• Basic data structure in python
• String object basics and inbuilt methods
• List: Object, methods, comprehensions
• Tuple: Object, methods, comprehensions
• Sets: Object, methods, comprehensions
• Dictionary: Object, methods, comprehensions

MODULE 4: PYTHON FUNCTIONS

• Functions basics
• Function Parameter passing
• Iterators
• Generator functions
• Lambda functions
• Map, reduce, filter functions

MODULE 5: PYTHON NUMPY PACKAGE

• NumPy Introduction
• Array –Data Structure
• Core Numpy functions
• Matrix Operations

MODULE 6: PYTHON PANDAS PACKAGE

• Pandas functions
• Data Frame and Series –Data Structure
• Data munging with Pandas
• Imputation and outlier analysis

MODULE 1: DATA SCIENCE ESSENTIALS

• Introduction to Data Science
• Data Science Terminologies
• Classifications of Analytics
• Data Science Project workflow

MODULE 2: DATA ENGINEERING FOUNDATION

• Introduction to Data Engineering
• Data engineering importance
• Ecosystems of data engineering tools
• Core concepts of data engineering

MODULE 3: PYTHON FOR DATA SCIENCE

• Introduction to Python
• Python Data Types, Operators
• Flow Control statements, Functions
• Structured vs Unstructured Data
• Python Numpy package introduction
• Array Data Structures in Numpy
• Array operations and methods
• Python Pandas package introduction
• Data Structures : Series and DataFrame
• Pandas DataFrame key methods

MODULE 4: VISUALIZATION WITH PYTHON

• Visualization Packages (Matplotlib)
• Components Of A Plot, Sub-Plots
• Basic Plots: Line, Bar, Pie, Scatter
• Advanced Python Data Visualizations

MODULE 5: R LANGUAGE ESSENTIALS

• R Installation and Setup
• R STUDIO –R Development Env
• R language basics and data structures
• R data structures , control statements

MODULE 6: STATISTICS

• Descriptive And Inferential statistics
• Types Of Data, Sampling types
• Measures of Central Tendencies
• Data Variability: Standard Deviation
• Z-Score, Outliers, Normal Distribution
• Central Limit Theorem
• Histogram, Normality Tests
• Skewness & Kurtosis
• Understanding Hypothesis Testing
• P-Value Method, Types Of Errors
• T Distribution, One Sample T-Test
• Independent And Relational T Tests
• Direct And Indirect Correlation
• Regression Theory

MODULE 7: MACHINE LEARNING INTRODUCTION

• Machine Learning Introduction
• ML core concepts
• Unsupervised and Supervised Learning
• Clustering with K-Means
• Regression and Classification Models.
• Regression Algorithm: Linear Regression
• ML Model Evaluation
• Classification Algorithm: Logistic Regression

MODULE 1: LINUX INTRODUCTION

• Introduction to Linux
• Shell Environment Basics
• Understanding Linux Kernel
• Distros in Linux
• Installing Linux in virtual box
• Linux Boot process
• Basic Linux commands

MODULE 2: LINUX SHELL SCRIPTING

• Shell scripting Introduction
• Setting shell script permission and execute
• Shell conditional statements
• IF, IF-ELSE and Nested IF statement
• Looping Statements: WHILE and FOR
• Functions in Shell script

MODULE 3: LINUX FILE MANAGEMENT

• Introduction to Linux file management
• Everything is a file in Linux (files, directories, executables and processes)
• Understanding Linux users, groups and processes, Root and Linux file hierarchy
• Understanding file permissions, CHMOD
• File copying, moving and deleting
• Process control commands (PS and KILL)
• Hand-on file management tasks

MODULE 4: SCHEDULING TASKS

• Introduction to Daemons
• Scheduling task in Linux
• Cron and Crontab
• Hands-on scheduling task in linux

MODULE 5: LINUX PACKAGE MANAGEMENT

• Package Management
• Package Managers & DPKG
• Working with APT & APT GET

MODULE 6: LINUX COMMANDS

• Part 1: sudo, pwd, cd, ls, cat, cp, mv, mkdir,
rmdir, rm, touch, locate, find, grep, df, du, head,
tail, diff, tar, chmod, chown, jobs, kill, ping

• Part 2: wget, uname, top, history, man, echo, zip,
unzip, hostname, useradd, userdel, apt-get,
nano, vi,jed,alias,unalias,su,htop

MODULE 7: DATABASE CONNECTIVITY

• Installing, configuring and security MySQL
• Executing SQL queries from the terminal
• Querying through shell script
• Running queries from a shell script
• Performing CRUD Operation
• Hands-on Exercise

MODULE 8: LINUX NETWORKING

• Networking in Linux
• Networking commands
• PING, IFCONFIG, Wget
• cURL,SSH, SCP and FTP, learning firewall tools: iptables
• firewalld, DSN and resolving IP adresss
• etc/hosts, etc/hostname, nslookup and dig

MODULE 9: PERMISSIONS & SECURITY

• Types of Account in Linux
• User Management, Group Management
• Files Access Controls, Linux File Permissions
• Modifying File Ownership
• Sudoers in Linux, Special Permissions
• System Management, System tools
• Hard link and Soft link, Aliasing in Linux
• Creating users in Multiple ways

MODULE 1: GIT INTRODUCTION

• Purpose of Version Control
• Popular Version control tools
• Git Distribution Version Control
• Terminologies
• Git Workflow
• Git Architecture

MODULE 2: GIT REPOSITORY and GitHub

• Git Repo Introduction
• Create New Repo with Init command
• Copying existing repo
• Git user and remote node
• Git Status and rebase
• Review Repo History
• GitHub Cloud Remote Repo

MODULE 3: COMMITS, PULL, FETCH AND PUSH

• Code commits
• Pull, Fetch and conflicts resolution
• Pushing to Remote Repo

MODULE 4: TAGGING, BRANCHING AND MERGING

• Organize code with branches
• Checkout branch
• Merge branches

MODULE 5: UNDOING CHANGES

• Editing Commits
• Commit command Amend flag
• Git reset and revert

MODULE 6: GIT WITH GITHUB AND BITBUCKET

• Creating GitHub Account
• Local and Remote Repo
• Collaborating with other developers
• Bitbucket Git account

MODULE 1: DVC INTRODUCTION

• DVC Purpose
• Managing Project with DVC
• DVC Workflow
• Tools for DVC version control

MODULE 2: GIT INTRODUCTION

• Purpose of Version Control
• Popular Version control tools
• Git Distribution Version Control
• Terminologies
• Git Workflow
• Git Architecture

MODULE 3: GIT REPOSITORY and GitHub

• Git Repo Introduction
• Create New Repo with Init command
• Copying existing repo
• Git user and remote node
• Git Status and rebase
• Review Repo History
• GitHub Cloud Remote Repo

MODULE 4: PYTHON DVC PACKAGE

• Python DV Installation
• Project folder setup
• DVC Configuration
• Integrating with Git Repo

MODULE 5: HANDS-ON DVC PROJECT

• Project Data
• DVC pipeline setup
• ML Modeling and Evaluation
• DVC Metrics
• Establishing repeated experiments with DVC

MODULE 1: AMAZON AWS DATA SERVICES

• Introduction to Linux
• Shell Environment Basics
• Understanding Linux Kernel
• Distros in Linux
• Installing Linux in virtual box
• Linux Boot process
• Basic Linux commands

MODULE 2: AWS MLOPS

• Setting shell script permission and execute
• Shell conditional statements
• IF, IF-ELSE and Nested IF statement
• Looping Statements: WHILE and FOR
• Functions in Shell script

MODULE 3: AZURE MLOPS

• Create an Azure machine learning workspace
• Setup a new project in Azure DevOps
• Import existing YAML pipeline to Azure DevOps
• Declare variables for CI/CD pipeline

MODULE 4: Azure ML Train & Deploy

• Create training compute
• Train ML model
• Register model
• Deploy model in AKS
• Hands-on: Build and Run MLOps

Machine learning is a subset of artificial intelligence that emphasizes a machine's ability to imitate intelligent human behavior. Artificial intelligence systems are used to complete complex tasks in just the same way that humans solve problems. Machine learning is a technique of AI that can be harnessed in a multitude of ways.

The discipline of delivering machine learning (ML) models through repeatable and efficient workflows is known as machine learning operations (MLOps).

Machine Learning Operations (MLOps) is an acronym for Machine Learning Operations. MLOps is a basic component of Machine Learning engineering that focuses on optimizing the process of deploying machine learning models, as well as maintaining and monitoring them.

The cyclical process that data science initiatives follow is known as the machine learning life cycle. It lays out each step that a company should take to generate tangible economic value from machine learning and artificial intelligence (AI).

Machine Learning's Mechanisms Machine learning uses two techniques: supervised learning, which entails training a model on known input and output data to predict future outputs, and unsupervised learning, which involves identifying interrelationships and patterns in input data.

MLOps intends to fast scale up machine learning ML model delivery in order to obtain corporate insights from data. Many companies have created a new profession called ML engineer to assure MLOps performance. 

Machine Learning Operations (MLOps) provides a technical backbone for managing the machine learning lifecycle through automation and scalability, allowing businesses to overcome many of the hurdles on the path to AI with ROI. AI and machine learning projects should influence your company's destiny.

• Obtaining information.

• Analyze data.

• Transformation and preparation of data.

• Training and development model

• Validation of the model

• Serving as a model.

• Observation of the model

• Model retraining 

From model generation (software development lifecycle, continuous integration/continuous delivery), orchestration, and deployment, to health, diagnostics, governance, and business KPIs, MLOps covers the complete lifecycle. MLOps is a subset of ModelOps, according to Gartner.

An end-to-end enterprise-grade platform for managing the complete machine learning and deep learning product life-cycle for data scientists, data engineers, DevOps, and managers. Data science is democratized with this AI platform, which automates end-to-end machine learning at scale.