DA 212-O MLOps at Scale 3:1 (Summer 2022)

Course Instructor: Prof. Sashikumaar Ganesan, CDS

Course description: This course is aimed to be an advaned graduate-level (200-series) course. This course is aimed at building the foundation of scalable (parallel) computing for Artificial Intelligence and Machine learning (AI & ML)

Syllabus

Parallel Computer Architecture: Pipelining and super-scalar processor, SIMD vectorization, Caches, Multicore architectures, GPUs, Data access optimization.

Programming Models: Shared Memory Programming basics, Shared memory programming with OpenMP, Message-passing, MPI, CUDA, MapReduce.

Machine Learning at Scale: Automatic parallelization with Numba, Dask, PySpark, Keras, Distributed training with TensorFlow, Deploying TensorFlow models in AWS.

MLOps: Introduction to MLOps, Foundations, MLOps for containers, Continuous Integration, Continuous Deployment for ML models, Monitoring and Feedback

Prerequisites: DA 203-O: Introduction to Computing for AI & Machine Learning (or) DA 224-O Practical Machine Learning (or) consent of Instructor

Textbooks / References

  1. Georg Hager, Gerhard Wellen. Introduction to High-Performance Computing for Scientists and Engineers. CRC Press, 2010.
  2. Noah Gift and Alfredo Deza: Practical MLOps, 1st Edition, O’Reilly Media, Inc., 2021
  3. Aurélien Géron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition, O'Reilly Media, Inc. 2019

Grading:

  • Assignments: 20
  • Midterm: 20
  • Project: 30
  • End Term: 30