Course Contents & Lec/Lab

Key Topics

  • Data collection, logging, and source instrumentation
  • Data validation and data augmentation
  • Dataset labeling, annotator agreement, and active learning
  • Hyperparameter search and AutoML frameworks
  • Model checkpointing and experiment tracking
  • Data version management
  • Docker containerization and cloud infrastructure (AWS/GCP/Azure)
  • Run-time profiling and memory footprint management
  • GPU utilization tracking and optimization
  • Production A/B testing and model redeployment
  • Covariate shift management and automated retraining
  • Model quantization and edge device deployment (TF-Lite)

Lecture and Lab Schedule

Lecture and lab topics and materials will be added as sessions are conducted.

DateSessionTopicMaterial
Aug 5 Lab 0 Intro to Kaggle Handout
Aug 6 Lecture 1 How to Collect Data Lec
Aug 12 Lab 1 Data Validation Lab Solution
Aug 13 Lecture 2 Methods of Data Validation Lec
Aug 17 Lab 2 Data Labelling Lab Solution
Aug 20 Lecture 3 Exercise on Data Collection & Validation Lec
Aug 24 Lab 3 Data Labelling (Cont.) Lab Solution
Aug 27 Lecture 4 Methods of Data Labelling Lec Exercise 2
Aug 31 Lab 4 Exp Management & Hyperparameter Search Lab
Sep 3 Lecture 5 Model and Code Reproducibility Lec
Sep 7 Lab 5 Exp Management & Hyperparameter Search Lab Solution
Sep 10 Lecture 6 Model and Code Reproducibility Lec Exercise 3
Sep 14 Lab 6 Checkpointing, Tracking, & Versioning Lab
Sep 17 Lecture 7 Review/Revision Session for Exam
Sep 24 Lecture 8 Exam Question Paper Discussion
Docker & Deployment
Question Paper Solution
Lec
Sep 28Lab 7
Oct 1Lecture 9
Oct 5Lab 8
Oct 8Lecture 10
Oct 12Lab 9
Oct 15Lecture 11
Oct 26Lab 10
Oct 29Lecture 12
Nov 2Lab 11
Nov 5Lecture 13
Nov 9Lab 12
Nov 12Lecture 14
Nov 16Lab 13
Nov 19Lecture 15