ML for Data Engineers

Course overview

The course aims to supply future Data Engineers with the basic concepts of designing and training Machine Learning models. The course reviews various algorithms of Supervised and Unsupervised conventional Machine Learning, as well as fundamentals of Deep Learning and Computer Vision. Real-world case studies accompany all the theoretical material, and at the end of the course, students have a chance to build and deploy a machine learning model for a real business problem.

Course topics

Topic 1: Intro

  • What is ML
  • A bit of history about ML
  • Supervised vs Unsupervised vs Reinforcement
  • ML Toolbox
  • ML Project Workflow

Topic 2: Conventional Supervised ML

  • Linear Regression
  • Non-linear Regression
  • Logistic Regression
  • Regularization
  • Intro to Decision Trees and GBMs

Topic 3: Conventional Unsupervised ML

  • Dimensionality Reduction
  • Eigenvalues
  • Clustering
  • Text Clustering
  • TF-IDF

Topic 4: Deep Learning and Intro to Computer Vision

  • Intro to Deep Learning
  • What Neural Network is?
  • What makes NN deep?
  • Inference
  • Backprop
  • Optimizers
  • Learning Rate

Topic 5: Finishing with Computer Vision and Brief Overview of Recommender Engines

  • Types of Computer Vision Problems
  • Detection
  • Segmentation
  • Classification
  • DL Runtimes
  • Model weights
  • TTA
  • Augmentations
  • Collaborative filtering
  • Recommender engines

Topic 6:  Deployment and monitoring of the ML models

  • How models could be exported
  • Validating the input
  • Most popular deployment options
  • Active Learning
  • Updating the model