Volume I: Foundations
Introduction to Machine Learning Systems
17 decks covering the full single-machine ML stack: from data engineering and neural network computation through training, compression, hardware acceleration, and deployment. 570 slides, 141 SVG figures, approximately 19 hours of teaching material.
| Ch | Title | Slides | SVGs | ~Time | Active Learning | PPTX | Source | |
|---|---|---|---|---|---|---|---|---|
| 0 | Course Overview | 27 | 5 | 53 min | 9 | PPTX | Source | |
| 1 | Introduction | 36 | 17 | 74 min | 9 | PPTX | Source | |
| 2 | ML Systems | 33 | 7 | 70 min | 8 | PPTX | Source | |
| 3 | ML Workflow | 36 | 8 | 79 min | 8 | PPTX | Source | |
| 4 | Data Engineering | 31 | 8 | 66 min | 9 | PPTX | Source | |
| 5 | Neural Network Computation | 33 | 7 | 68 min | 9 | PPTX | Source | |
| 6 | Network Architectures | 30 | 7 | 63 min | 9 | PPTX | Source | |
| 7 | ML Frameworks | 34 | 7 | 73 min | 8 | PPTX | Source | |
| 8 | Model Training | 34 | 7 | 68 min | 8 | PPTX | Source | |
| 9 | Data Selection | 30 | 7 | 63 min | 8 | PPTX | Source | |
| 10 | Model Compression | 30 | 9 | 64 min | 8 | PPTX | Source | |
| 11 | Hardware Acceleration | 31 | 7 | 66 min | 9 | PPTX | Source | |
| 12 | Benchmarking | 32 | 9 | 68 min | 8 | PPTX | Source | |
| 13 | Model Serving | 35 | 7 | 74 min | 10 | PPTX | Source | |
| 14 | ML Operations | 33 | 6 | 69 min | 9 | PPTX | Source | |
| 15 | Responsible Engineering | 30 | 5 | 61 min | 8 | PPTX | Source | |
| 16 | Conclusion | 25 | 5 | 51 min | 8 | PPTX | Source | |
| Total | 570 | 141 | ~19 hrs | 145 |
Tip
PPTX files are image-based (300 DPI) — visually identical to the PDF. Use them for PowerPoint presenter mode and slide annotations. For editable slides, download the LaTeX source.