This is Christian Mills' personal website.
Introduction The Symptoms of an Unstructured AI Project How a Roadmap Creates Predictability and Focus From Hopeful Experiment to Engineering Investment Introduction In the early stages of an AI project, the inherent uncertainty can make ‘playing it by ear’ feel like the right move. In their desire to verify a project’s feasibility, teams often over-focus on the initial proof-of-concept at the…
Introduction Diagnosing the Data Bottleneck: Why “More Data” Isn’t a Strategy The Solution: From Data Collection to Data Manufacturing A ‘Data Factory’ Isn’t a Magic Bullet—It’s an Engine That Needs a Driver Is a Synthetic Data Approach Right for Your Project? Introduction There’s a predictable point where many AI projects stall: when they move from curated test samples to the real world. The…
Introduction Part 3: Knowledge Part 2: Reasoning Part 1: Language Structures Tip Resource Links YouTube Recording: ICML 2024 Tutorial: Physics of Language Models Project Page: Physics of Language Models Speaker: Zeyuan Allen-Zhu Introduction Spectrum of “Theory” in Language Models The term “theory” in the context of language models encompasses a broad spectrum, ranging from rigorous mathematical…
Epigraph Introduction Chapter 1: We Will Not Inflict Daily Standups on Our Devs Chapter 2: We Will Not Test Devs with Computer Science Riddles Chapter 3: We Will Not Recruit 10x Developers Chapter 4: We Will Not Let Devs Start without an Estimate Chapter 5: We Will Not Sprint Chapter 6: We Will Not Allow Our Devs to Multitask Chapter 7: We Will Not Accept the First Solution a Dev Thinks Up Chapter…
Introduction Physics Chemistry Biology Afterthoughts Note Book LInks: Series Homepage Introduction Overview Purpose and Scope Core Premise : The world is complex and mysterious, but humans have worked some of its fundamental principles. Main Goal : Present a “latticework of mental models” for approaching: New ideas Situations Problems Challenges Series Structure Volume 1 : Covered nine general…
Tip This post is part of the following series: GPU MODE Lecture Notes : My notes from the GPU MODE reading group lectures run by Andreas Kopf and Mark Saroufim . Introduction Notation Clarifications Tensors in Cutlass Layout and Coordinate Systems Sub-Tiles and Memory Organization Tiling CuTe Framework Practical Implementation and Usage Tip Resource Links: YouTube Recording: Lecture 15: CUTLASS…
Tip This post is part of the following series: GPU MODE Lecture Notes : My notes from the GPU MODE reading group lectures run by Andreas Kopf and Mark Saroufim . Introduction Overview of the Talk Why and When to Use Triton How to Write Triton Kernels Practical Examples Benchmarking Auto-Tuning Conclusion and Resources Tip Resource Links: YouTube Recording: Lecture 14: Practitioners Guide to Triton…
Tip This post is part of the following series: Fine-Tuning Image Classifiers with PyTorch and the timm library for Beginners Introduction Quantization Process Getting Started with the Code Setting Up Your Python Environment Importing the Required Dependencies Setting Up the Project Loading the Checkpoint Data Loading the Dataset Collecting Calibration Data Performing Inference with TensorRT…
Tip This post is part of the following series: Training YOLOX Models for Real-Time Object Detection in PyTorch Introduction Quantization Process Getting Started with the Code Setting Up Your Python Environment Importing the Required Dependencies Setting Up the Project Loading the Checkpoint Data Loading the Dataset Collecting Calibration Data Performing Inference with TensorRT Conclusion…
Tip This post is part of the following series: GPU MODE Lecture Notes : My notes from the GPU MODE reading group lectures run by Andreas Kopf and Mark Saroufim . Introduction and Overview Motivation: Long Context Transformers and Applications Background: Multimodal Input and Transformers Challenge: Memory Limitations Vanilla Attention Recap Compute Scaling with Context Length Online Softmax and…
Part I: Entering Through the Gates of Insight Chapter 1: Hunting for Insights Chapter 2: The Flash of Illumination Chapter 3: Connections Chapter 4: Coincidences and Curiosities Chapter 5: Contradictions Chapter 6: Creative Desperation Chapter 7: Different Ways to Look at Insight Chapter 8: The Logic of Discovery Part II: Shutting the Gates Chapter 9: Stupidity Chapter 10: The Study of Contrasting…
Tip This post is part of the following series: Training YOLOX Models for Real-Time Object Detection in PyTorch Introduction Prerequisites Compiling Our Model Preparing the Raspberry Pi Setting Up a Python Virtual Environment Getting Started with the Code Importing the Required Dependencies Define Utility Functions Setting Up the Project Loading the Inference Resources Tracking Objects in a Camera…
Introduction An Overview of Machine Learning, PyTorch, and Jupyter Further Learning Installing Mamba Access the Miniforge Prompt Benefits of Using Virtual Environments Creating a Python Environment Package Overview Installing PyTorch and Jupyter Testing PyTorch and Jupyter Notebook Setup Managing and Updating Packages with Mamba Conclusion Introduction This tutorial will guide you through…
Preface Introduction: Acquiring Wisdom Mental Model #1: The Map is Not the Territory Mental Model #2: Circle of Confidence Mental Model #3: First Principles Thinking Mental Model #4: Thought Experiment Mental Model #5: Second-Order Thinking Mental Model #6: Probabilistic Thinking Mental Model #7: Inversion Mental Model #8: Occam’s Razor Mental Model #9: Hanlon’s Razor Note Book LInks: Series…
Introduction: Let’s write a new story together Part I꞉ Rise of the AI Tutor Part II: Giving Voice to the Social Sciences Part III: Empowering the Next Innovators Part IV: Better Together Part V: Keeping Kids Safe Part VI: Teaching in the Age of AI Part VII: The Global Classroom Part VIII: AI Assessments and Admissions Part IX: Work and What Comes Next Note Book LInks: Publisher Page Introduction:…
Introduction by Ann Druyan Preface: My Teachers Chapter 1: The Most Precious Thing Chapter 2: Science and Hope Chapter 3: The Man in the Moon and the Face on Mars Chapter 4: Aliens Chapter 5: Spoofing and Secrecy Chapter 6: Hallucinations Chapter 7: The Demon-Haunted World Chapter 8: On the Distinction Between True and False Visions Chapter 9: Therapy Chapter 10: The Dragon in My Garage Chapter…
Introduction Enable Resizable BAR in BIOS Install Ubuntu Verify Resizable BAR Install Drivers Set Up a Python Environment Modify PyTorch Code Local LLM Inference with IPEX-LLM Conclusion Introduction In this tutorial, I’ll guide you through setting up Intel’s PyTorch extension on Ubuntu to train models with their Arc GPUs . The extension provides Intel’s latest feature optimizations and hardware…
Introduction Enable Resizable BAR in BIOS Install Drivers Install Microsoft Visual C++ Redistributable Disable Integrated Graphics Set Up a Python Environment Modify PyTorch Code Conclusion Introduction In this tutorial, I’ll guide you through setting up Intel’s PyTorch extension on Windows to train models with their Arc GPUs . The extension provides Intel’s latest feature optimizations and…
Tip This post is part of the following series: GPU MODE Lecture Notes : My notes from the GPU MODE reading group lectures run by Andreas Kopf and Mark Saroufim . Introduction Background: The Need for Optimization Memory Hierarchy and Performance Attention Mechanisms: A Brief Overview Flash Attention: Tiling Strategy The Softmax Challenge Flash Attention 2: Advanced Techniques Implementation…
Tip This post is part of the following series: GPU MODE Lecture Notes : My notes from the GPU MODE reading group lectures run by Andreas Kopf and Mark Saroufim . Introduction The Performance Challenge of Sparsity Unstructured Sparsity and Its Limitations on GPUs Semi-structured Sparsity (2:4 Sparsity) Block Sparsity Composing Sparsity with Quantization Sparse Training Future Directions and…