Full Stack ML- Python Fundamentals to Advanced MLOps

Description

Course Outline

Module 1: Python Programming Essentials (15 Hours)

• Objective: Build a strong Python foundation, necessary for efficient machine learning and data handling.
• Topics:
  • Python fundamentals: syntax, data types, operators, and control flow
  • Data structures: lists, dictionaries, sets, and tuples
  • Functions, error handling, and file operations
  • Introduction to Object-Oriented Programming (OOP)
  • Libraries for data analysis and visualization: NumPy, Pandas, Matplotlib
  • Hands-on Project: Data processing and visualization mini-project

Module 2: Introduction to Machine Learning (20 Hours)

• Objective: Introduce core machine learning concepts and foundational algorithms.
• Topics:
  • Basics of machine learning and types of learning (supervised vs. unsupervised)
  • Data preprocessing: data cleaning, scaling, and encoding techniques
  • Regression algorithms: linear regression, logistic regression
  • Classification techniques: decision trees, k-nearest neighbors
  • Clustering: k-means clustering, hierarchical clustering
  • Model evaluation metrics: accuracy, precision, recall, F1-score
  • Hands-on Project: Build and evaluate a machine learning model using Scikit-Learn

Module 3: Advanced Machine Learning Techniques (15 Hours)

• Objective: Build expertise in complex models, feature engineering, and neural networks.
• Topics:
  • Ensemble methods: random forests, boosting (AdaBoost, XGBoost)
  • Dimensionality reduction: PCA and t-SNE
  • Introduction to deep learning and neural networks
  • Neural network architecture, activation functions, forward and backpropagation
  • Introduction to transfer learning and pre-trained models
  • Hands-on Project: Develop and train a simple neural network using Keras/TensorFlow

Module 4: Advanced MLOps for Model Deployment and Management (25 Hours)

• Objective: Develop operational skills to deploy, manage, and monitor machine learning models in production.
• Topics:
  • MLOps, DevOps, and DataOps Pipelines

• Overview of MLOps, DevOps, and DataOps in ML lifecycle management
• Pipeline building for continuous integration and deployment (CI/CD) in ML workflows
• Hands-on: Implement a CI/CD pipeline using GitHub Actions or Jenkins

• Experiment Tracking and Model Versioning
  • Using MLflow for experiment tracking, model versioning, and deployment
  • Importance of reproducibility and managing multiple model versions
  • Hands-on: Track and manage model training experiments with MLflow
• Model Deployment with Flask/FastAPI
  • Building REST APIs for model inference using Flask or FastAPI
  • Post-deployment monitoring and managing model performance
  • Hands-on: Deploy a trained model with Flask/FastAPI and monitor key metrics
• Cloud-Based MLOps with Amazon SageMaker and Azure ML
  • Amazon SageMaker: Streamlining model training, deployment, and monitoring
    • Hands-on: Deploy a model on Amazon SageMaker and explore automated monitoring
  • Azure ML: Model training and deployment using Microsoft’s Azure platform
    • Hands-on: Deploy and manage a model using Azure ML
• Advanced Tools: Hugging Face for NLP
  • Introduction to Hugging Face Transformers and Model Hub for NLP tasks
  • Fine-tuning pre-trained models and deploying NLP models
  • Hands-on: Fine-tune and deploy an NLP model using Hugging Face
• Continuous Monitoring and Model Retraining
  • Monitoring deployed models for performance and drift
  • Automating model retraining in response to data drift
  • Continuous learning pipelines for adaptive models
  • Capstone Project: Build a full MLOps pipeline using MLflow, cloud deployment (Amazon SageMaker/Azure ML), and monitoring

Course Time: 

• Total 40 Class, Per Class 2 Hours (Online Mode) 

Course Fee

• 6000 BDT (60 USD)

Instructors

For Python and Machine Learning

Md. Mossaddek Touhid

Lecturer

Northern University, Khulna.

Bsc in CSE from KUET

For MLOps

Md. Hasan Ali

3+ years of Industry Experience

AI Engineer, WSD

Ex. ML and Data Engineer at Brainstation23

BSc in Software Engineering from SUST