github-starred/TinyTorch
2
0
Fork 0
mirror of https://github.com/MLSysBook/TinyTorch.git synced 2026-05-23 19:35:56 -05:00
Code Issues 7 Packages Projects Releases Wiki Activity
Files
gh-pages
TinyTorch/dev/instructor-guide.html
github-actions[bot] c5e8a28d75 Deploy dev site to /dev/ from dev branch - 4035bfcb42 4035bfcb42
2025-11-25 18:08:28 +00:00

1476 lines
97 KiB
HTML
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
<!DOCTYPE html>
<html lang="en" data-content_root="./" >
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
<title>👩‍🏫 TinyTorch Instructor Guide &#8212; Tiny🔥Torch</title>
<script data-cfasync="false">
document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
</script>
<!-- Loaded before other Sphinx assets -->
<link href="_static/styles/theme.css?digest=dfe6caa3a7d634c4db9b" rel="stylesheet" />
<link href="_static/styles/bootstrap.css?digest=dfe6caa3a7d634c4db9b" rel="stylesheet" />
<link href="_static/styles/pydata-sphinx-theme.css?digest=dfe6caa3a7d634c4db9b" rel="stylesheet" />
<link href="_static/vendor/fontawesome/6.5.2/css/all.min.css?digest=dfe6caa3a7d634c4db9b" rel="stylesheet" />
<link rel="preload" as="font" type="font/woff2" crossorigin href="_static/vendor/fontawesome/6.5.2/webfonts/fa-solid-900.woff2" />
<link rel="preload" as="font" type="font/woff2" crossorigin href="_static/vendor/fontawesome/6.5.2/webfonts/fa-brands-400.woff2" />
<link rel="preload" as="font" type="font/woff2" crossorigin href="_static/vendor/fontawesome/6.5.2/webfonts/fa-regular-400.woff2" />
<link rel="stylesheet" type="text/css" href="_static/pygments.css?v=03e43079" />
<link rel="stylesheet" type="text/css" href="_static/styles/sphinx-book-theme.css?v=eba8b062" />
<link rel="stylesheet" type="text/css" href="_static/togglebutton.css?v=13237357" />
<link rel="stylesheet" type="text/css" href="_static/copybutton.css?v=76b2166b" />
<link rel="stylesheet" type="text/css" href="_static/mystnb.8ecb98da25f57f5357bf6f572d296f466b2cfe2517ffebfabe82451661e28f02.css" />
<link rel="stylesheet" type="text/css" href="_static/sphinx-thebe.css?v=4fa983c6" />
<link rel="stylesheet" type="text/css" href="_static/sphinx-design.min.css?v=95c83b7e" />
<link rel="stylesheet" type="text/css" href="_static/custom.css?v=afcf7c3c" />
<!-- Pre-loaded scripts that we'll load fully later -->
<link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=dfe6caa3a7d634c4db9b" />
<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=dfe6caa3a7d634c4db9b" />
<script src="_static/vendor/fontawesome/6.5.2/js/all.min.js?digest=dfe6caa3a7d634c4db9b"></script>
<script src="_static/documentation_options.js?v=9eb32ce0"></script>
<script src="_static/doctools.js?v=9a2dae69"></script>
<script src="_static/sphinx_highlight.js?v=dc90522c"></script>
<script src="_static/clipboard.min.js?v=a7894cd8"></script>
<script src="_static/copybutton.js?v=f281be69"></script>
<script src="_static/scripts/sphinx-book-theme.js?v=887ef09a"></script>
<script>let toggleHintShow = 'Click to show';</script>
<script>let toggleHintHide = 'Click to hide';</script>
<script>let toggleOpenOnPrint = 'true';</script>
<script src="_static/togglebutton.js?v=4a39c7ea"></script>
<script>var togglebuttonSelector = '.toggle, .admonition.dropdown';</script>
<script src="_static/design-tabs.js?v=f930bc37"></script>
<script>const THEBE_JS_URL = "https://unpkg.com/thebe@0.8.2/lib/index.js"; const thebe_selector = ".thebe,.cell"; const thebe_selector_input = "pre"; const thebe_selector_output = ".output, .cell_output"</script>
<script async="async" src="_static/sphinx-thebe.js?v=c100c467"></script>
<script>var togglebuttonSelector = '.toggle, .admonition.dropdown';</script>
<script>const THEBE_JS_URL = "https://unpkg.com/thebe@0.8.2/lib/index.js"; const thebe_selector = ".thebe,.cell"; const thebe_selector_input = "pre"; const thebe_selector_output = ".output, .cell_output"</script>
<script>DOCUMENTATION_OPTIONS.pagename = 'instructor-guide';</script>
<script src="_static/ml-timeline.js?v=76e9b3e3"></script>
<script src="_static/wip-banner.js?v=5357532b"></script>
<script src="_static/marimo-badges.js?v=1e5d2842"></script>
<script src="_static/sidebar-link.js?v=404b701b"></script>
<script src="_static/hero-carousel.js?v=10341d2a"></script>
<link rel="icon" href="_static/favicon.svg"/>
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="Teaching Assistant Guide for TinyTorch" href="usage-paths/ta-guide.html" />
<link rel="prev" title="TinyTorch for Instructors: Complete ML Systems Course" href="usage-paths/classroom-use.html" />
<meta name="viewport" content="width=device-width, initial-scale=1"/>
<meta name="docsearch:language" content="en"/>
</head>
<body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
<div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
<div id="pst-scroll-pixel-helper"></div>
<button type="button" class="btn rounded-pill" id="pst-back-to-top">
<i class="fa-solid fa-arrow-up"></i>Back to top</button>
<input type="checkbox"
class="sidebar-toggle"
id="pst-primary-sidebar-checkbox"/>
<label class="overlay overlay-primary" for="pst-primary-sidebar-checkbox"></label>
<input type="checkbox"
class="sidebar-toggle"
id="pst-secondary-sidebar-checkbox"/>
<label class="overlay overlay-secondary" for="pst-secondary-sidebar-checkbox"></label>
<div class="search-button__wrapper">
<div class="search-button__overlay"></div>
<div class="search-button__search-container">
<form class="bd-search d-flex align-items-center"
action="search.html"
method="get">
<i class="fa-solid fa-magnifying-glass"></i>
<input type="search"
class="form-control"
name="q"
id="search-input"
placeholder="Search this book..."
aria-label="Search this book..."
autocomplete="off"
autocorrect="off"
autocapitalize="off"
spellcheck="false"/>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
</form></div>
</div>
<div class="pst-async-banner-revealer d-none">
<aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
</div>
<header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
</header>
<div class="bd-container">
<div class="bd-container__inner bd-page-width">
<div class="bd-sidebar-primary bd-sidebar">
<div class="sidebar-header-items sidebar-primary__section">
</div>
<div class="sidebar-primary-items__start sidebar-primary__section">
<div class="sidebar-primary-item">
<a class="navbar-brand logo" href="intro.html">
<img src="_static/logo-tinytorch.png" class="logo__image only-light" alt="Tiny🔥Torch - Home"/>
<script>document.write(`<img src="_static/logo-tinytorch.png" class="logo__image only-dark" alt="Tiny🔥Torch - Home"/>`);</script>
</a></div>
<div class="sidebar-primary-item">
<script>
document.write(`
<button class="btn search-button-field search-button__button" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass"></i>
<span class="search-button__default-text">Search</span>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
</button>
`);
</script></div>
<div class="sidebar-primary-item"><nav class="bd-links bd-docs-nav" aria-label="Main">
<div class="bd-toc-item navbar-nav active">
<ul class="nav bd-sidenav bd-sidenav__home-link">
<li class="toctree-l1">
<a class="reference internal" href="intro.html">
Getting Started
</a>
</li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">🚀 Getting Started</span></p>
<ul class="current nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="quickstart-guide.html">Quick Start Guide</a></li>
<li class="toctree-l1"><a class="reference internal" href="student-workflow.html">Student Workflow</a></li>
<li class="toctree-l1"><a class="reference internal" href="usage-paths/classroom-use.html">For Instructors</a></li>
<li class="toctree-l1 current active"><a class="current reference internal" href="#">Instructor Guide</a></li>
<li class="toctree-l1"><a class="reference internal" href="usage-paths/ta-guide.html">TA Guide</a></li>
<li class="toctree-l1"><a class="reference internal" href="usage-paths/team-onboarding.html">Team Onboarding</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">🏗 Foundation Tier (01-07)</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="tiers/foundation.html">📖 Tier Overview</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/01_tensor_ABOUT.html">01. Tensor</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/02_activations_ABOUT.html">02. Activations</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/03_layers_ABOUT.html">03. Layers</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/04_losses_ABOUT.html">04. Losses</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/05_autograd_ABOUT.html">05. Autograd</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/06_optimizers_ABOUT.html">06. Optimizers</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/07_training_ABOUT.html">07. Training</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">🏛️ Architecture Tier (08-13)</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="tiers/architecture.html">📖 Tier Overview</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/08_dataloader_ABOUT.html">08. DataLoader</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/09_spatial_ABOUT.html">09. Convolutions</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/10_tokenization_ABOUT.html">10. Tokenization</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/11_embeddings_ABOUT.html">11. Embeddings</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/12_attention_ABOUT.html">12. Attention</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/13_transformers_ABOUT.html">13. Transformers</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">⏱️ Optimization Tier (14-19)</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="tiers/optimization.html">📖 Tier Overview</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/14_profiling_ABOUT.html">14. Profiling</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/15_quantization_ABOUT.html">15. Quantization</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/16_compression_ABOUT.html">16. Compression</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/17_memoization_ABOUT.html">17. Memoization</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/18_acceleration_ABOUT.html">18. Acceleration</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/19_benchmarking_ABOUT.html">19. Benchmarking</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">🏅 Capstone Competition</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="tiers/olympics.html">📖 Competition Overview</a></li>
<li class="toctree-l1"><a class="reference internal" href="modules/20_capstone_ABOUT.html">20. Torch Olympics</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">🧭 Course Orientation</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="chapters/00-introduction.html">Course Structure</a></li>
<li class="toctree-l1"><a class="reference internal" href="prerequisites.html">Prerequisites &amp; Resources</a></li>
<li class="toctree-l1"><a class="reference internal" href="chapters/learning-journey.html">Learning Journey</a></li>
<li class="toctree-l1"><a class="reference internal" href="chapters/milestones.html">Historical Milestones</a></li>
<li class="toctree-l1"><a class="reference internal" href="faq.html">FAQ</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">🛠️ TITO CLI Reference</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="tito/overview.html">Command Overview</a></li>
<li class="toctree-l1"><a class="reference internal" href="tito/modules.html">Module Workflow</a></li>
<li class="toctree-l1"><a class="reference internal" href="tito/milestones.html">Milestone System</a></li>
<li class="toctree-l1"><a class="reference internal" href="tito/data.html">Progress &amp; Data</a></li>
<li class="toctree-l1"><a class="reference internal" href="tito/troubleshooting.html">Troubleshooting</a></li>
<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets Guide</a></li>
</ul>
<p aria-level="2" class="caption" role="heading"><span class="caption-text">🤝 Community</span></p>
<ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="community.html">Ecosystem</a></li>
<li class="toctree-l1"><a class="reference internal" href="resources.html">Learning Resources</a></li>
<li class="toctree-l1"><a class="reference internal" href="credits.html">Credits &amp; Acknowledgments</a></li>
</ul>
</div>
</nav></div>
</div>
<div class="sidebar-primary-items__end sidebar-primary__section">
</div>
<div id="rtd-footer-container"></div>
</div>
<main id="main-content" class="bd-main" role="main">
<div class="sbt-scroll-pixel-helper"></div>
<div class="bd-content">
<div class="bd-article-container">
<div class="bd-header-article d-print-none">
<div class="header-article-items header-article__inner">
<div class="header-article-items__start">
<div class="header-article-item"><button class="sidebar-toggle primary-toggle btn btn-sm" title="Toggle primary sidebar" data-bs-placement="bottom" data-bs-toggle="tooltip">
<span class="fa-solid fa-bars"></span>
</button></div>
</div>
<div class="header-article-items__end">
<div class="header-article-item">
<div class="article-header-buttons">
<div class="dropdown dropdown-source-buttons">
<button class="btn dropdown-toggle" type="button" data-bs-toggle="dropdown" aria-expanded="false" aria-label="Source repositories">
<i class="fab fa-github"></i>
</button>
<ul class="dropdown-menu">
<li><a href="https://github.com/mlsysbook/TinyTorch" target="_blank"
class="btn btn-sm btn-source-repository-button dropdown-item"
title="Source repository"
data-bs-placement="left" data-bs-toggle="tooltip"
>
<span class="btn__icon-container">
<i class="fab fa-github"></i>
</span>
<span class="btn__text-container">Repository</span>
</a>
</li>
<li><a href="https://github.com/mlsysbook/TinyTorch/edit/main/site/instructor-guide.md" target="_blank"
class="btn btn-sm btn-source-edit-button dropdown-item"
title="Suggest edit"
data-bs-placement="left" data-bs-toggle="tooltip"
>
<span class="btn__icon-container">
<i class="fas fa-pencil-alt"></i>
</span>
<span class="btn__text-container">Suggest edit</span>
</a>
</li>
<li><a href="https://github.com/mlsysbook/TinyTorch/issues/new?title=Issue%20on%20page%20%2Finstructor-guide.html&body=Your%20issue%20content%20here." target="_blank"
class="btn btn-sm btn-source-issues-button dropdown-item"
title="Open an issue"
data-bs-placement="left" data-bs-toggle="tooltip"
>
<span class="btn__icon-container">
<i class="fas fa-lightbulb"></i>
</span>
<span class="btn__text-container">Open issue</span>
</a>
</li>
</ul>
</div>
<div class="dropdown dropdown-download-buttons">
<button class="btn dropdown-toggle" type="button" data-bs-toggle="dropdown" aria-expanded="false" aria-label="Download this page">
<i class="fas fa-download"></i>
</button>
<ul class="dropdown-menu">
<li><a href="_sources/instructor-guide.md" target="_blank"
class="btn btn-sm btn-download-source-button dropdown-item"
title="Download source file"
data-bs-placement="left" data-bs-toggle="tooltip"
>
<span class="btn__icon-container">
<i class="fas fa-file"></i>
</span>
<span class="btn__text-container">.md</span>
</a>
</li>
<li>
<button onclick="window.print()"
class="btn btn-sm btn-download-pdf-button dropdown-item"
title="Print to PDF"
data-bs-placement="left" data-bs-toggle="tooltip"
>
<span class="btn__icon-container">
<i class="fas fa-file-pdf"></i>
</span>
<span class="btn__text-container">.pdf</span>
</button>
</li>
</ul>
</div>
<button onclick="toggleFullScreen()"
class="btn btn-sm btn-fullscreen-button"
title="Fullscreen mode"
data-bs-placement="bottom" data-bs-toggle="tooltip"
>
<span class="btn__icon-container">
<i class="fas fa-expand"></i>
</span>
</button>
<script>
document.write(`
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button" title="light/dark" aria-label="light/dark" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="theme-switch fa-solid fa-sun fa-lg" data-mode="light"></i>
<i class="theme-switch fa-solid fa-moon fa-lg" data-mode="dark"></i>
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"></i>
</button>
`);
</script>
<script>
document.write(`
<button class="btn btn-sm pst-navbar-icon search-button search-button__button" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass fa-lg"></i>
</button>
`);
</script>
<button class="sidebar-toggle secondary-toggle btn btn-sm" title="Toggle secondary sidebar" data-bs-placement="bottom" data-bs-toggle="tooltip">
<span class="fa-solid fa-list"></span>
</button>
</div></div>
</div>
</div>
</div>
<div id="jb-print-docs-body" class="onlyprint">
<h1>👩‍🏫 TinyTorch Instructor Guide</h1>
<!-- Table of contents -->
<div id="print-main-content">
<div id="jb-print-toc">
<div>
<h2> Contents </h2>
</div>
<nav aria-label="Page">
<ul class="visible nav section-nav flex-column">
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#course-overview">🎯 Course Overview</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#instructor-setup">🛠️ Instructor Setup</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#initial-setup"><strong>1. Initial Setup</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#verify-installation"><strong>2. Verify Installation</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#assignment-workflow">📝 Assignment Workflow</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#simplified-with-tito-cli"><strong>Simplified with Tito CLI</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#prepare-assignments"><strong>1. Prepare Assignments</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#distribute-to-students"><strong>2. Distribute to Students</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#collect-submissions"><strong>3. Collect Submissions</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#auto-grade"><strong>4. Auto-Grade</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#manual-review"><strong>5. Manual Review</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#generate-feedback"><strong>6. Generate Feedback</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#export-grades"><strong>7. Export Grades</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#grading-components">📊 Grading Components</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#auto-graded-70"><strong>Auto-Graded (70%)</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#manually-graded-30"><strong>Manually Graded (30%)</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#grading-rubric-for-ml-systems-questions"><strong>Grading Rubric for ML Systems Questions</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#sample-solutions-for-grading-calibration">📋 Sample Solutions for Grading Calibration</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-01-tensor-memory-footprint">Module 01: Tensor - Memory Footprint</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-05-autograd-backward-pass">Module 05: Autograd - Backward Pass</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-09-spatial-convolution-implementation">Module 09: Spatial - Convolution Implementation</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-12-attention-scaled-dot-product-attention">Module 12: Attention - Scaled Dot-Product Attention</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#grading-guidelines-using-sample-solutions">Grading Guidelines Using Sample Solutions</a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-teaching-notes">📚 Module Teaching Notes</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-01-tensor"><strong>Module 01: Tensor</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-02-activations"><strong>Module 02: Activations</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-04-05-layers-networks"><strong>Module 04-05: Layers &amp; Networks</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-06-07-spatial-attention"><strong>Module 06-07: Spatial &amp; Attention</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-08-11-training-pipeline"><strong>Module 08-11: Training Pipeline</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-12-15-production"><strong>Module 12-15: Production</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-16-tinygpt"><strong>Module 16: TinyGPT</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#learning-objectives">🎯 Learning Objectives</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#tracking-progress">📈 Tracking Progress</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#individual-progress"><strong>Individual Progress</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#class-overview"><strong>Class Overview</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#identify-struggling-students"><strong>Identify Struggling Students</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#teaching-tips">💡 Teaching Tips</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#emphasize-building-over-theory"><strong>1. Emphasize Building Over Theory</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#connect-to-production-systems"><strong>2. Connect to Production Systems</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#make-performance-visible"><strong>3. Make Performance Visible</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#encourage-systems-questions"><strong>4. Encourage Systems Questions</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#troubleshooting">🔧 Troubleshooting</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#common-student-issues"><strong>Common Student Issues</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#nbgrader-issues"><strong>NBGrader Issues</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#sample-schedule-16-weeks">📊 Sample Schedule (16 Weeks)</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#assessment-strategy">🎓 Assessment Strategy</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#continuous-assessment-70"><strong>Continuous Assessment (70%)</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#projects-30"><strong>Projects (30%)</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#additional-resources">📚 Additional Resources</a></li>
</ul>
</nav>
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section id="tinytorch-instructor-guide">
<h1>👩‍🏫 TinyTorch Instructor Guide<a class="headerlink" href="#tinytorch-instructor-guide" title="Link to this heading">#</a></h1>
<p>Complete guide for teaching ML Systems Engineering with TinyTorch.</p>
<section id="course-overview">
<h2>🎯 Course Overview<a class="headerlink" href="#course-overview" title="Link to this heading">#</a></h2>
<p>TinyTorch teaches ML systems engineering through building, not just using. Students construct a complete ML framework from tensors to transformers, understanding memory, performance, and scaling at each step.</p>
</section>
<section id="instructor-setup">
<h2>🛠️ Instructor Setup<a class="headerlink" href="#instructor-setup" title="Link to this heading">#</a></h2>
<section id="initial-setup">
<h3><strong>1. Initial Setup</strong><a class="headerlink" href="#initial-setup" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Clone and setup</span>
git<span class="w"> </span>clone<span class="w"> </span>https://github.com/MLSysBook/TinyTorch.git
<span class="nb">cd</span><span class="w"> </span>TinyTorch
<span class="c1"># Virtual environment (MANDATORY)</span>
python<span class="w"> </span>-m<span class="w"> </span>venv<span class="w"> </span>.venv
<span class="nb">source</span><span class="w"> </span>.venv/bin/activate
<span class="c1"># Install with instructor tools</span>
pip<span class="w"> </span>install<span class="w"> </span>-r<span class="w"> </span>requirements.txt
pip<span class="w"> </span>install<span class="w"> </span>nbgrader
<span class="c1"># Setup grading infrastructure</span>
tito<span class="w"> </span>grade<span class="w"> </span>setup
</pre></div>
</div>
</section>
<section id="verify-installation">
<h3><strong>2. Verify Installation</strong><a class="headerlink" href="#verify-installation" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>tito<span class="w"> </span>system<span class="w"> </span>doctor
<span class="c1"># Should show all green checkmarks</span>
tito<span class="w"> </span>grade
<span class="c1"># Should show available grade commands</span>
</pre></div>
</div>
</section>
</section>
<section id="assignment-workflow">
<h2>📝 Assignment Workflow<a class="headerlink" href="#assignment-workflow" title="Link to this heading">#</a></h2>
<section id="simplified-with-tito-cli">
<h3><strong>Simplified with Tito CLI</strong><a class="headerlink" href="#simplified-with-tito-cli" title="Link to this heading">#</a></h3>
<p>We’ve wrapped NBGrader behind simple <code class="docutils literal notranslate"><span class="pre">tito</span> <span class="pre">grade</span></code> commands so you don’t need to learn NBGrader’s complex interface.</p>
</section>
<section id="prepare-assignments">
<h3><strong>1. Prepare Assignments</strong><a class="headerlink" href="#prepare-assignments" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Generate instructor version (with solutions)</span>
tito<span class="w"> </span>grade<span class="w"> </span>generate<span class="w"> </span>01_tensor
<span class="c1"># Create student version (solutions removed)</span>
tito<span class="w"> </span>grade<span class="w"> </span>release<span class="w"> </span>01_tensor
<span class="c1"># Student version will be in: release/tinytorch/01_tensor/</span>
</pre></div>
</div>
</section>
<section id="distribute-to-students">
<h3><strong>2. Distribute to Students</strong><a class="headerlink" href="#distribute-to-students" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Option A: GitHub Classroom (recommended)</span>
<span class="c1"># 1. Create assignment repository from TinyTorch</span>
<span class="c1"># 2. Remove solutions from modules</span>
<span class="c1"># 3. Students clone and work</span>
<span class="c1"># Option B: Direct distribution</span>
<span class="c1"># Share the release/ directory contents</span>
</pre></div>
</div>
</section>
<section id="collect-submissions">
<h3><strong>3. Collect Submissions</strong><a class="headerlink" href="#collect-submissions" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Collect all students</span>
tito<span class="w"> </span>grade<span class="w"> </span>collect<span class="w"> </span>01_tensor
<span class="c1"># Or specific student</span>
tito<span class="w"> </span>grade<span class="w"> </span>collect<span class="w"> </span>01_tensor<span class="w"> </span>--student<span class="w"> </span>student_id
</pre></div>
</div>
</section>
<section id="auto-grade">
<h3><strong>4. Auto-Grade</strong><a class="headerlink" href="#auto-grade" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Grade all submissions</span>
tito<span class="w"> </span>grade<span class="w"> </span>autograde<span class="w"> </span>01_tensor
<span class="c1"># Grade specific student</span>
tito<span class="w"> </span>grade<span class="w"> </span>autograde<span class="w"> </span>01_tensor<span class="w"> </span>--student<span class="w"> </span>student_id
</pre></div>
</div>
</section>
<section id="manual-review">
<h3><strong>5. Manual Review</strong><a class="headerlink" href="#manual-review" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Open grading interface (browser-based)</span>
tito<span class="w"> </span>grade<span class="w"> </span>manual<span class="w"> </span>01_tensor
<span class="c1"># This launches a web interface for:</span>
<span class="c1"># - Reviewing ML Systems question responses</span>
<span class="c1"># - Adding feedback comments</span>
<span class="c1"># - Adjusting auto-grades</span>
</pre></div>
</div>
</section>
<section id="generate-feedback">
<h3><strong>6. Generate Feedback</strong><a class="headerlink" href="#generate-feedback" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Create feedback files for students</span>
tito<span class="w"> </span>grade<span class="w"> </span>feedback<span class="w"> </span>01_tensor
</pre></div>
</div>
</section>
<section id="export-grades">
<h3><strong>7. Export Grades</strong><a class="headerlink" href="#export-grades" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Export all grades to CSV</span>
tito<span class="w"> </span>grade<span class="w"> </span><span class="nb">export</span>
<span class="c1"># Or specific module</span>
tito<span class="w"> </span>grade<span class="w"> </span><span class="nb">export</span><span class="w"> </span>--module<span class="w"> </span>01_tensor<span class="w"> </span>--output<span class="w"> </span>grades_module01.csv
</pre></div>
</div>
</section>
</section>
<section id="grading-components">
<h2>📊 Grading Components<a class="headerlink" href="#grading-components" title="Link to this heading">#</a></h2>
<section id="auto-graded-70">
<h3><strong>Auto-Graded (70%)</strong><a class="headerlink" href="#auto-graded-70" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p>Code implementation correctness</p></li>
<li><p>Test passing</p></li>
<li><p>Function signatures</p></li>
<li><p>Output validation</p></li>
</ul>
</section>
<section id="manually-graded-30">
<h3><strong>Manually Graded (30%)</strong><a class="headerlink" href="#manually-graded-30" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p>ML Systems Thinking questions (3 per module)</p></li>
<li><p>Each question: 10 points</p></li>
<li><p>Focus on understanding, not perfection</p></li>
</ul>
</section>
<section id="grading-rubric-for-ml-systems-questions">
<h3><strong>Grading Rubric for ML Systems Questions</strong><a class="headerlink" href="#grading-rubric-for-ml-systems-questions" title="Link to this heading">#</a></h3>
<div class="pst-scrollable-table-container"><table class="table">
<thead>
<tr class="row-odd"><th class="head"><p>Points</p></th>
<th class="head"><p>Criteria</p></th>
</tr>
</thead>
<tbody>
<tr class="row-even"><td><p>9-10</p></td>
<td><p>Demonstrates deep understanding, references specific code, discusses systems implications</p></td>
</tr>
<tr class="row-odd"><td><p>7-8</p></td>
<td><p>Good understanding, some code references, basic systems thinking</p></td>
</tr>
<tr class="row-even"><td><p>5-6</p></td>
<td><p>Surface understanding, generic response, limited systems perspective</p></td>
</tr>
<tr class="row-odd"><td><p>3-4</p></td>
<td><p>Attempted but misses key concepts</p></td>
</tr>
<tr class="row-even"><td><p>0-2</p></td>
<td><p>No attempt or completely off-topic</p></td>
</tr>
</tbody>
</table>
</div>
<p><strong>What to Look For:</strong></p>
<ul class="simple">
<li><p>References to actual implemented code</p></li>
<li><p>Memory/performance analysis</p></li>
<li><p>Scaling considerations</p></li>
<li><p>Production system comparisons</p></li>
<li><p>Understanding of trade-offs</p></li>
</ul>
</section>
</section>
<section id="sample-solutions-for-grading-calibration">
<h2>📋 Sample Solutions for Grading Calibration<a class="headerlink" href="#sample-solutions-for-grading-calibration" title="Link to this heading">#</a></h2>
<p>This section provides sample solutions to help calibrate grading standards. Use these as reference points when evaluating student submissions.</p>
<section id="module-01-tensor-memory-footprint">
<h3>Module 01: Tensor - Memory Footprint<a class="headerlink" href="#module-01-tensor-memory-footprint" title="Link to this heading">#</a></h3>
<p><strong>Excellent Solution (9-10 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">memory_footprint</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="w"> </span><span class="sd">&quot;&quot;&quot;Calculate tensor memory in bytes.&quot;&quot;&quot;</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">nbytes</span>
</pre></div>
</div>
<p><strong>Why Excellent</strong>:</p>
<ul class="simple">
<li><p>Concise and correct</p></li>
<li><p>Uses NumPy’s built-in <code class="docutils literal notranslate"><span class="pre">nbytes</span></code> property</p></li>
<li><p>Clear docstring</p></li>
<li><p>Handles all tensor shapes correctly</p></li>
</ul>
<p><strong>Good Solution (7-8 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">memory_footprint</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="w"> </span><span class="sd">&quot;&quot;&quot;Calculate memory usage.&quot;&quot;&quot;</span>
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">prod</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">dtype</span><span class="o">.</span><span class="n">itemsize</span>
</pre></div>
</div>
<p><strong>Why Good</strong>:</p>
<ul class="simple">
<li><p>Correct implementation</p></li>
<li><p>Manually calculates (shows understanding)</p></li>
<li><p>Works but less efficient than using <code class="docutils literal notranslate"><span class="pre">nbytes</span></code></p></li>
<li><p>Minor: docstring could be more specific</p></li>
</ul>
<p><strong>Acceptable Solution (5-6 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">memory_footprint</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="n">size</span> <span class="o">=</span> <span class="mi">1</span>
<span class="k">for</span> <span class="n">dim</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">shape</span><span class="p">:</span>
<span class="n">size</span> <span class="o">*=</span> <span class="n">dim</span>
<span class="k">return</span> <span class="n">size</span> <span class="o">*</span> <span class="mi">4</span> <span class="c1"># Assumes float32</span>
</pre></div>
</div>
<p><strong>Why Acceptable</strong>:</p>
<ul class="simple">
<li><p>Correct logic but hardcoded dtype size</p></li>
<li><p>Works for float32 but fails for other dtypes</p></li>
<li><p>Shows understanding of memory calculation</p></li>
<li><p>Missing proper dtype handling</p></li>
</ul>
</section>
<section id="module-05-autograd-backward-pass">
<h3>Module 05: Autograd - Backward Pass<a class="headerlink" href="#module-05-autograd-backward-pass" title="Link to this heading">#</a></h3>
<p><strong>Excellent Solution (9-10 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">backward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">gradient</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
<span class="w"> </span><span class="sd">&quot;&quot;&quot;Backward pass through computational graph.&quot;&quot;&quot;</span>
<span class="k">if</span> <span class="n">gradient</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">gradient</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">ones_like</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">grad</span> <span class="o">=</span> <span class="n">gradient</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
<span class="c1"># Compute gradients for inputs</span>
<span class="n">input_grads</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="o">.</span><span class="n">backward</span><span class="p">(</span><span class="n">gradient</span><span class="p">)</span>
<span class="c1"># Propagate to input tensors</span>
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">input_grads</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">):</span>
<span class="k">for</span> <span class="n">input_tensor</span><span class="p">,</span> <span class="n">input_grad</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="o">.</span><span class="n">inputs</span><span class="p">,</span> <span class="n">input_grads</span><span class="p">):</span>
<span class="k">if</span> <span class="n">input_tensor</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">:</span>
<span class="n">input_tensor</span><span class="o">.</span><span class="n">backward</span><span class="p">(</span><span class="n">input_grad</span><span class="p">)</span>
<span class="k">else</span><span class="p">:</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="o">.</span><span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">:</span>
<span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="o">.</span><span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">backward</span><span class="p">(</span><span class="n">input_grads</span><span class="p">)</span>
</pre></div>
</div>
<p><strong>Why Excellent</strong>:</p>
<ul class="simple">
<li><p>Handles both scalar and tensor gradients</p></li>
<li><p>Properly checks <code class="docutils literal notranslate"><span class="pre">requires_grad</span></code> before propagating</p></li>
<li><p>Handles tuple returns from grad_fn</p></li>
<li><p>Clear variable names and structure</p></li>
</ul>
<p><strong>Good Solution (7-8 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">backward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">gradient</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
<span class="k">if</span> <span class="n">gradient</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">gradient</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">ones_like</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
<span class="bp">self</span><span class="o">.</span><span class="n">grad</span> <span class="o">=</span> <span class="n">gradient</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="p">:</span>
<span class="n">grads</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="o">.</span><span class="n">backward</span><span class="p">(</span><span class="n">gradient</span><span class="p">)</span>
<span class="k">for</span> <span class="n">inp</span><span class="p">,</span> <span class="n">grad</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="o">.</span><span class="n">inputs</span><span class="p">,</span> <span class="n">grads</span><span class="p">):</span>
<span class="n">inp</span><span class="o">.</span><span class="n">backward</span><span class="p">(</span><span class="n">grad</span><span class="p">)</span>
</pre></div>
</div>
<p><strong>Why Good</strong>:</p>
<ul class="simple">
<li><p>Correct logic</p></li>
<li><p>Missing <code class="docutils literal notranslate"><span class="pre">requires_grad</span></code> check (minor issue)</p></li>
<li><p>Assumes grads is always iterable (may fail for single input)</p></li>
<li><p>Works for most cases but less robust</p></li>
</ul>
<p><strong>Acceptable Solution (5-6 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">backward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">grad</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">grad</span> <span class="o">=</span> <span class="n">grad</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="p">:</span>
<span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="o">.</span><span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">backward</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">grad_fn</span><span class="o">.</span><span class="n">backward</span><span class="p">(</span><span class="n">grad</span><span class="p">))</span>
</pre></div>
</div>
<p><strong>Why Acceptable</strong>:</p>
<ul class="simple">
<li><p>Basic backward pass works</p></li>
<li><p>Only handles single input (fails for multi-input operations)</p></li>
<li><p>Missing None gradient handling</p></li>
<li><p>Shows understanding but incomplete</p></li>
</ul>
</section>
<section id="module-09-spatial-convolution-implementation">
<h3>Module 09: Spatial - Convolution Implementation<a class="headerlink" href="#module-09-spatial-convolution-implementation" title="Link to this heading">#</a></h3>
<p><strong>Excellent Solution (9-10 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
<span class="w"> </span><span class="sd">&quot;&quot;&quot;Forward pass with explicit loops for clarity.&quot;&quot;&quot;</span>
<span class="n">batch_size</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
<span class="n">out_height</span> <span class="o">=</span> <span class="p">(</span><span class="n">height</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span> <span class="o">+</span> <span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">)</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">+</span> <span class="mi">1</span>
<span class="n">out_width</span> <span class="o">=</span> <span class="p">(</span><span class="n">width</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span> <span class="o">+</span> <span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">)</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">+</span> <span class="mi">1</span>
<span class="n">output</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="n">batch_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">out_height</span><span class="p">,</span> <span class="n">out_width</span><span class="p">))</span>
<span class="c1"># Apply padding</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
<span class="n">x</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="p">((</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">),</span>
<span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">)),</span> <span class="n">mode</span><span class="o">=</span><span class="s1">&#39;constant&#39;</span><span class="p">)</span>
<span class="c1"># Explicit convolution loops</span>
<span class="k">for</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">batch_size</span><span class="p">):</span>
<span class="k">for</span> <span class="n">oc</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">):</span>
<span class="k">for</span> <span class="n">oh</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">out_height</span><span class="p">):</span>
<span class="k">for</span> <span class="n">ow</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">out_width</span><span class="p">):</span>
<span class="n">h_start</span> <span class="o">=</span> <span class="n">oh</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span>
<span class="n">w_start</span> <span class="o">=</span> <span class="n">ow</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span>
<span class="n">h_end</span> <span class="o">=</span> <span class="n">h_start</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span>
<span class="n">w_end</span> <span class="o">=</span> <span class="n">w_start</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span>
<span class="n">window</span> <span class="o">=</span> <span class="n">x</span><span class="p">[</span><span class="n">b</span><span class="p">,</span> <span class="p">:,</span> <span class="n">h_start</span><span class="p">:</span><span class="n">h_end</span><span class="p">,</span> <span class="n">w_start</span><span class="p">:</span><span class="n">w_end</span><span class="p">]</span>
<span class="n">output</span><span class="p">[</span><span class="n">b</span><span class="p">,</span> <span class="n">oc</span><span class="p">,</span> <span class="n">oh</span><span class="p">,</span> <span class="n">ow</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span>
<span class="n">window</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="p">[</span><span class="n">oc</span><span class="p">]</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">bias</span><span class="p">[</span><span class="n">oc</span><span class="p">]</span>
<span class="p">)</span>
<span class="k">return</span> <span class="n">Tensor</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">requires_grad</span><span class="o">=</span><span class="n">x</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">)</span>
</pre></div>
</div>
<p><strong>Why Excellent</strong>:</p>
<ul class="simple">
<li><p>Clear output shape calculation</p></li>
<li><p>Proper padding handling</p></li>
<li><p>Explicit loops make O(kernel_size²) complexity visible</p></li>
<li><p>Correct gradient tracking setup</p></li>
<li><p>Well-structured and readable</p></li>
</ul>
<p><strong>Good Solution (7-8 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
<span class="n">B</span><span class="p">,</span> <span class="n">C</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
<span class="n">out_h</span> <span class="o">=</span> <span class="p">(</span><span class="n">H</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">)</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">+</span> <span class="mi">1</span>
<span class="n">out_w</span> <span class="o">=</span> <span class="p">(</span><span class="n">W</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">)</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">+</span> <span class="mi">1</span>
<span class="n">out</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="n">B</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">out_h</span><span class="p">,</span> <span class="n">out_w</span><span class="p">))</span>
<span class="k">for</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">B</span><span class="p">):</span>
<span class="k">for</span> <span class="n">oc</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">):</span>
<span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">out_h</span><span class="p">):</span>
<span class="k">for</span> <span class="n">j</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">out_w</span><span class="p">):</span>
<span class="n">h</span> <span class="o">=</span> <span class="n">i</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span>
<span class="n">w</span> <span class="o">=</span> <span class="n">j</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span>
<span class="n">out</span><span class="p">[</span><span class="n">b</span><span class="p">,</span> <span class="n">oc</span><span class="p">,</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span>
<span class="n">x</span><span class="p">[</span><span class="n">b</span><span class="p">,</span> <span class="p">:,</span> <span class="n">h</span><span class="p">:</span><span class="n">h</span><span class="o">+</span><span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span><span class="n">w</span><span class="o">+</span><span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">]</span>
<span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="p">[</span><span class="n">oc</span><span class="p">]</span>
<span class="p">)</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">bias</span><span class="p">[</span><span class="n">oc</span><span class="p">]</span>
<span class="k">return</span> <span class="n">Tensor</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
</pre></div>
</div>
<p><strong>Why Good</strong>:</p>
<ul class="simple">
<li><p>Correct implementation</p></li>
<li><p>Missing padding support (works only for padding=0)</p></li>
<li><p>Less clear variable names</p></li>
<li><p>Missing requires_grad propagation</p></li>
</ul>
<p><strong>Acceptable Solution (5-6 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
<span class="n">out</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span><span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span><span class="o">-</span><span class="mi">2</span><span class="p">))</span>
<span class="k">for</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]):</span>
<span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">):</span>
<span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">out</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">]):</span>
<span class="k">for</span> <span class="n">j</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">out</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">3</span><span class="p">]):</span>
<span class="n">out</span><span class="p">[</span><span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">,</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="n">b</span><span class="p">,</span> <span class="p">:,</span> <span class="n">i</span><span class="p">:</span><span class="n">i</span><span class="o">+</span><span class="mi">3</span><span class="p">,</span> <span class="n">j</span><span class="p">:</span><span class="n">j</span><span class="o">+</span><span class="mi">3</span><span class="p">]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="p">[</span><span class="n">c</span><span class="p">])</span>
<span class="k">return</span> <span class="n">Tensor</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
</pre></div>
</div>
<p><strong>Why Acceptable</strong>:</p>
<ul class="simple">
<li><p>Basic convolution works</p></li>
<li><p>Hardcoded kernel_size=3 (not general)</p></li>
<li><p>No stride or padding support</p></li>
<li><p>Shows understanding but incomplete</p></li>
</ul>
</section>
<section id="module-12-attention-scaled-dot-product-attention">
<h3>Module 12: Attention - Scaled Dot-Product Attention<a class="headerlink" href="#module-12-attention-scaled-dot-product-attention" title="Link to this heading">#</a></h3>
<p><strong>Excellent Solution (9-10 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">query</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
<span class="w"> </span><span class="sd">&quot;&quot;&quot;Scaled dot-product attention with numerical stability.&quot;&quot;&quot;</span>
<span class="c1"># Compute attention scores</span>
<span class="n">scores</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">query</span><span class="p">,</span> <span class="n">key</span><span class="o">.</span><span class="n">T</span><span class="p">)</span> <span class="o">/</span> <span class="n">np</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">d_k</span><span class="p">)</span>
<span class="c1"># Apply mask if provided</span>
<span class="k">if</span> <span class="n">mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">scores</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">mask</span><span class="p">,</span> <span class="n">scores</span><span class="p">,</span> <span class="o">-</span><span class="mf">1e9</span><span class="p">)</span>
<span class="c1"># Softmax with numerical stability</span>
<span class="n">exp_scores</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">scores</span> <span class="o">-</span> <span class="n">np</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">scores</span><span class="p">,</span> <span class="n">axis</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span> <span class="n">keepdims</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
<span class="n">attention_weights</span> <span class="o">=</span> <span class="n">exp_scores</span> <span class="o">/</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">exp_scores</span><span class="p">,</span> <span class="n">axis</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span> <span class="n">keepdims</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="c1"># Apply attention to values</span>
<span class="n">output</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">attention_weights</span><span class="p">,</span> <span class="n">value</span><span class="p">)</span>
<span class="k">return</span> <span class="n">output</span><span class="p">,</span> <span class="n">attention_weights</span>
</pre></div>
</div>
<p><strong>Why Excellent</strong>:</p>
<ul class="simple">
<li><p>Proper scaling factor (1/√d_k)</p></li>
<li><p>Numerical stability with max subtraction</p></li>
<li><p>Mask handling</p></li>
<li><p>Returns both output and attention weights</p></li>
<li><p>Clear and well-documented</p></li>
</ul>
<p><strong>Good Solution (7-8 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">):</span>
<span class="n">scores</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="o">.</span><span class="n">T</span><span class="p">)</span> <span class="o">/</span> <span class="n">np</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">q</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">])</span>
<span class="n">weights</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">scores</span><span class="p">)</span> <span class="o">/</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">scores</span><span class="p">),</span> <span class="n">axis</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span> <span class="n">keepdims</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">weights</span><span class="p">,</span> <span class="n">v</span><span class="p">)</span>
</pre></div>
</div>
<p><strong>Why Good</strong>:</p>
<ul class="simple">
<li><p>Correct implementation</p></li>
<li><p>Missing numerical stability (may overflow)</p></li>
<li><p>Missing mask support</p></li>
<li><p>Works but less robust</p></li>
</ul>
<p><strong>Acceptable Solution (5-6 points)</strong>:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">):</span>
<span class="n">scores</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="o">.</span><span class="n">T</span><span class="p">)</span>
<span class="n">weights</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">scores</span><span class="p">)</span> <span class="o">/</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">scores</span><span class="p">))</span>
<span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">weights</span><span class="p">,</span> <span class="n">v</span><span class="p">)</span>
</pre></div>
</div>
<p><strong>Why Acceptable</strong>:</p>
<ul class="simple">
<li><p>Basic attention mechanism</p></li>
<li><p>Missing scaling factor</p></li>
<li><p>Missing numerical stability</p></li>
<li><p>Incorrect softmax (should be per-row)</p></li>
</ul>
</section>
<section id="grading-guidelines-using-sample-solutions">
<h3>Grading Guidelines Using Sample Solutions<a class="headerlink" href="#grading-guidelines-using-sample-solutions" title="Link to this heading">#</a></h3>
<p><strong>When Evaluating Student Code</strong>:</p>
<ol class="arabic simple">
<li><p><strong>Correctness First</strong>: Does it pass all tests?</p>
<ul class="simple">
<li><p>If no: Maximum 6 points (even if well-written)</p></li>
<li><p>If yes: Proceed to quality evaluation</p></li>
</ul>
</li>
<li><p><strong>Code Quality</strong>:</p>
<ul class="simple">
<li><p><strong>Excellent (9-10)</strong>: Production-ready, handles edge cases, well-documented</p></li>
<li><p><strong>Good (7-8)</strong>: Correct and functional, minor improvements possible</p></li>
<li><p><strong>Acceptable (5-6)</strong>: Works but incomplete or has issues</p></li>
</ul>
</li>
<li><p><strong>Systems Thinking</strong>:</p>
<ul class="simple">
<li><p><strong>Excellent</strong>: Discusses memory, performance, scaling implications</p></li>
<li><p><strong>Good</strong>: Some systems awareness</p></li>
<li><p><strong>Acceptable</strong>: Focuses only on correctness</p></li>
</ul>
</li>
<li><p><strong>Common Patterns</strong>:</p>
<ul class="simple">
<li><p>Look for: Proper error handling, edge case consideration, documentation</p></li>
<li><p>Red flags: Hardcoded values, missing checks, unclear variable names</p></li>
</ul>
</li>
</ol>
<p><strong>Remember</strong>: These are calibration examples. Adjust based on your course level and learning objectives. The goal is consistent evaluation, not perfection.</p>
</section>
</section>
<section id="module-teaching-notes">
<h2>📚 Module Teaching Notes<a class="headerlink" href="#module-teaching-notes" title="Link to this heading">#</a></h2>
<section id="module-01-tensor">
<h3><strong>Module 01: Tensor</strong><a class="headerlink" href="#module-01-tensor" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p><strong>Focus</strong>: Memory layout, data structures</p></li>
<li><p><strong>Key Concept</strong>: Understanding memory is crucial for ML performance</p></li>
<li><p><strong>Demo</strong>: Show memory profiling, copying behavior</p></li>
</ul>
</section>
<section id="module-02-activations">
<h3><strong>Module 02: Activations</strong><a class="headerlink" href="#module-02-activations" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p><strong>Focus</strong>: Vectorization, numerical stability</p></li>
<li><p><strong>Key Concept</strong>: Small details matter at scale</p></li>
<li><p><strong>Demo</strong>: Gradient vanishing/exploding</p></li>
</ul>
</section>
<section id="module-04-05-layers-networks">
<h3><strong>Module 04-05: Layers &amp; Networks</strong><a class="headerlink" href="#module-04-05-layers-networks" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p><strong>Focus</strong>: Composition, parameter management</p></li>
<li><p><strong>Key Concept</strong>: Building blocks combine into complex systems</p></li>
<li><p><strong>Project</strong>: Build a small CNN</p></li>
</ul>
</section>
<section id="module-06-07-spatial-attention">
<h3><strong>Module 06-07: Spatial &amp; Attention</strong><a class="headerlink" href="#module-06-07-spatial-attention" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p><strong>Focus</strong>: Algorithmic complexity, memory patterns</p></li>
<li><p><strong>Key Concept</strong>: O(N²) operations become bottlenecks</p></li>
<li><p><strong>Demo</strong>: Profile attention memory usage</p></li>
</ul>
</section>
<section id="module-08-11-training-pipeline">
<h3><strong>Module 08-11: Training Pipeline</strong><a class="headerlink" href="#module-08-11-training-pipeline" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p><strong>Focus</strong>: End-to-end system integration</p></li>
<li><p><strong>Key Concept</strong>: Many components must work together</p></li>
<li><p><strong>Project</strong>: Train a real model</p></li>
</ul>
</section>
<section id="module-12-15-production">
<h3><strong>Module 12-15: Production</strong><a class="headerlink" href="#module-12-15-production" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p><strong>Focus</strong>: Deployment, optimization, monitoring</p></li>
<li><p><strong>Key Concept</strong>: Academic vs production requirements</p></li>
<li><p><strong>Demo</strong>: Model compression, deployment</p></li>
</ul>
</section>
<section id="module-16-tinygpt">
<h3><strong>Module 16: TinyGPT</strong><a class="headerlink" href="#module-16-tinygpt" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p><strong>Focus</strong>: Framework generalization</p></li>
<li><p><strong>Key Concept</strong>: 70% component reuse from vision to language</p></li>
<li><p><strong>Capstone</strong>: Build a working language model</p></li>
</ul>
</section>
</section>
<section id="learning-objectives">
<h2>🎯 Learning Objectives<a class="headerlink" href="#learning-objectives" title="Link to this heading">#</a></h2>
<p>By course end, students should be able to:</p>
<ol class="arabic simple">
<li><p><strong>Build</strong> complete ML systems from scratch</p></li>
<li><p><strong>Analyze</strong> memory usage and computational complexity</p></li>
<li><p><strong>Debug</strong> performance bottlenecks</p></li>
<li><p><strong>Optimize</strong> for production deployment</p></li>
<li><p><strong>Understand</strong> framework design decisions</p></li>
<li><p><strong>Apply</strong> systems thinking to ML problems</p></li>
</ol>
</section>
<section id="tracking-progress">
<h2>📈 Tracking Progress<a class="headerlink" href="#tracking-progress" title="Link to this heading">#</a></h2>
<section id="individual-progress">
<h3><strong>Individual Progress</strong><a class="headerlink" href="#individual-progress" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Check specific student progress</span>
tito<span class="w"> </span>checkpoint<span class="w"> </span>status<span class="w"> </span>--student<span class="w"> </span>student_id
</pre></div>
</div>
</section>
<section id="class-overview">
<h3><strong>Class Overview</strong><a class="headerlink" href="#class-overview" title="Link to this heading">#</a></h3>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Export all checkpoint achievements</span>
tito<span class="w"> </span>checkpoint<span class="w"> </span><span class="nb">export</span><span class="w"> </span>--output<span class="w"> </span>class_progress.csv
</pre></div>
</div>
</section>
<section id="identify-struggling-students">
<h3><strong>Identify Struggling Students</strong><a class="headerlink" href="#identify-struggling-students" title="Link to this heading">#</a></h3>
<p>Look for:</p>
<ul class="simple">
<li><p>Missing checkpoint achievements</p></li>
<li><p>Low scores on ML Systems questions</p></li>
<li><p>Incomplete module submissions</p></li>
</ul>
</section>
</section>
<section id="teaching-tips">
<h2>💡 Teaching Tips<a class="headerlink" href="#teaching-tips" title="Link to this heading">#</a></h2>
<section id="emphasize-building-over-theory">
<h3><strong>1. Emphasize Building Over Theory</strong><a class="headerlink" href="#emphasize-building-over-theory" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p>Have students type every line of code</p></li>
<li><p>Run tests immediately after implementation</p></li>
<li><p>Break and fix things intentionally</p></li>
</ul>
</section>
<section id="connect-to-production-systems">
<h3><strong>2. Connect to Production Systems</strong><a class="headerlink" href="#connect-to-production-systems" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p>Show PyTorch/TensorFlow equivalents</p></li>
<li><p>Discuss real-world bottlenecks</p></li>
<li><p>Share production war stories</p></li>
</ul>
</section>
<section id="make-performance-visible">
<h3><strong>3. Make Performance Visible</strong><a class="headerlink" href="#make-performance-visible" title="Link to this heading">#</a></h3>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Use profilers liberally</span>
<span class="k">with</span> <span class="n">TimeProfiler</span><span class="p">(</span><span class="s2">&quot;operation&quot;</span><span class="p">):</span>
<span class="n">result</span> <span class="o">=</span> <span class="n">expensive_operation</span><span class="p">()</span>
<span class="c1"># Show memory usage</span>
<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Memory: </span><span class="si">{</span><span class="n">get_memory_usage</span><span class="p">()</span><span class="si">:</span><span class="s2">.2f</span><span class="si">}</span><span class="s2"> MB&quot;</span><span class="p">)</span>
</pre></div>
</div>
</section>
<section id="encourage-systems-questions">
<h3><strong>4. Encourage Systems Questions</strong><a class="headerlink" href="#encourage-systems-questions" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p>“What would break at 1B parameters?”</p></li>
<li><p>“How would you distributed this?”</p></li>
<li><p>“What’s the bottleneck here?”</p></li>
</ul>
</section>
</section>
<section id="troubleshooting">
<h2>🔧 Troubleshooting<a class="headerlink" href="#troubleshooting" title="Link to this heading">#</a></h2>
<section id="common-student-issues">
<h3><strong>Common Student Issues</strong><a class="headerlink" href="#common-student-issues" title="Link to this heading">#</a></h3>
<p><strong>Environment Problems</strong></p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Student fix:</span>
tito<span class="w"> </span>system<span class="w"> </span>doctor
tito<span class="w"> </span>system<span class="w"> </span>reset
</pre></div>
</div>
<p><strong>Module Import Errors</strong></p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Rebuild package</span>
tito<span class="w"> </span><span class="nb">export</span><span class="w"> </span>--all
</pre></div>
</div>
<p><strong>Test Failures</strong></p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Detailed test output</span>
tito<span class="w"> </span>module<span class="w"> </span><span class="nb">test</span><span class="w"> </span>MODULE<span class="w"> </span>--verbose
</pre></div>
</div>
</section>
<section id="nbgrader-issues">
<h3><strong>NBGrader Issues</strong><a class="headerlink" href="#nbgrader-issues" title="Link to this heading">#</a></h3>
<p><strong>Database Locked</strong></p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Clear NBGrader database</span>
rm<span class="w"> </span>gradebook.db
tito<span class="w"> </span>grade<span class="w"> </span>setup
</pre></div>
</div>
<p><strong>Missing Submissions</strong></p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># Check submission directory</span>
ls<span class="w"> </span>submitted/*/MODULE/
</pre></div>
</div>
</section>
</section>
<section id="sample-schedule-16-weeks">
<h2>📊 Sample Schedule (16 Weeks)<a class="headerlink" href="#sample-schedule-16-weeks" title="Link to this heading">#</a></h2>
<div class="pst-scrollable-table-container"><table class="table">
<thead>
<tr class="row-odd"><th class="head"><p>Week</p></th>
<th class="head"><p>Module</p></th>
<th class="head"><p>Focus</p></th>
</tr>
</thead>
<tbody>
<tr class="row-even"><td><p>1</p></td>
<td><p>01 Tensor</p></td>
<td><p>Data Structures, Memory</p></td>
</tr>
<tr class="row-odd"><td><p>2</p></td>
<td><p>02 Activations</p></td>
<td><p>Non-linearity Functions</p></td>
</tr>
<tr class="row-even"><td><p>3</p></td>
<td><p>03 Layers</p></td>
<td><p>Neural Network Components</p></td>
</tr>
<tr class="row-odd"><td><p>4</p></td>
<td><p>04 Losses</p></td>
<td><p>Optimization Objectives</p></td>
</tr>
<tr class="row-even"><td><p>5</p></td>
<td><p>05 Autograd</p></td>
<td><p>Automatic Differentiation</p></td>
</tr>
<tr class="row-odd"><td><p>6</p></td>
<td><p>06 Optimizers</p></td>
<td><p>Training Algorithms</p></td>
</tr>
<tr class="row-even"><td><p>7</p></td>
<td><p>07 Training</p></td>
<td><p>Complete Training Loop</p></td>
</tr>
<tr class="row-odd"><td><p>8</p></td>
<td><p>Midterm Project</p></td>
<td><p>Build and Train Network</p></td>
</tr>
<tr class="row-even"><td><p>9</p></td>
<td><p>08 DataLoader</p></td>
<td><p>Data Pipeline</p></td>
</tr>
<tr class="row-odd"><td><p>10</p></td>
<td><p>09 Spatial</p></td>
<td><p>Convolutions, CNNs</p></td>
</tr>
<tr class="row-even"><td><p>11</p></td>
<td><p>10 Tokenization</p></td>
<td><p>Text Processing</p></td>
</tr>
<tr class="row-odd"><td><p>12</p></td>
<td><p>11 Embeddings</p></td>
<td><p>Word Representations</p></td>
</tr>
<tr class="row-even"><td><p>13</p></td>
<td><p>12 Attention</p></td>
<td><p>Attention Mechanisms</p></td>
</tr>
<tr class="row-odd"><td><p>14</p></td>
<td><p>13 Transformers</p></td>
<td><p>Transformer Architecture</p></td>
</tr>
<tr class="row-even"><td><p>15</p></td>
<td><p>14-19 Optimization</p></td>
<td><p>Profiling, Quantization, etc.</p></td>
</tr>
<tr class="row-odd"><td><p>16</p></td>
<td><p>20 Capstone</p></td>
<td><p>Torch Olympics Competition</p></td>
</tr>
</tbody>
</table>
</div>
</section>
<section id="assessment-strategy">
<h2>🎓 Assessment Strategy<a class="headerlink" href="#assessment-strategy" title="Link to this heading">#</a></h2>
<section id="continuous-assessment-70">
<h3><strong>Continuous Assessment (70%)</strong><a class="headerlink" href="#continuous-assessment-70" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p>Module completion: 4% each × 16 = 64%</p></li>
<li><p>Checkpoint achievements: 6%</p></li>
</ul>
</section>
<section id="projects-30">
<h3><strong>Projects (30%)</strong><a class="headerlink" href="#projects-30" title="Link to this heading">#</a></h3>
<ul class="simple">
<li><p>Midterm: Build and train CNN (15%)</p></li>
<li><p>Final: Extend TinyGPT (15%)</p></li>
</ul>
</section>
</section>
<section id="additional-resources">
<h2>📚 Additional Resources<a class="headerlink" href="#additional-resources" title="Link to this heading">#</a></h2>
<ul class="simple">
<li><p><a class="reference external" href="https://mlsysbook.ai">MLSys Book</a> - Companion textbook</p></li>
<li><p><a class="reference external" href="https://github.com/MLSysBook/TinyTorch/discussions">Course Discussions</a></p></li>
<li><p><a class="reference external" href="https://github.com/MLSysBook/TinyTorch/issues">Issue Tracker</a></p></li>
</ul>
<hr class="docutils" />
<p><strong>Need help? Open an issue or contact the TinyTorch team!</strong></p>
</section>
</section>
<script type="text/x-thebe-config">
{
requestKernel: true,
binderOptions: {
repo: "binder-examples/jupyter-stacks-datascience",
ref: "master",
},
codeMirrorConfig: {
theme: "abcdef",
mode: "python"
},
kernelOptions: {
name: "python3",
path: "./."
},
predefinedOutput: true
}
</script>
<script>kernelName = 'python3'</script>
</article>
<footer class="prev-next-footer d-print-none">
<div class="prev-next-area">
<a class="left-prev"
href="usage-paths/classroom-use.html"
title="previous page">
<i class="fa-solid fa-angle-left"></i>
<div class="prev-next-info">
<p class="prev-next-subtitle">previous</p>
<p class="prev-next-title">TinyTorch for Instructors: Complete ML Systems Course</p>
</div>
</a>
<a class="right-next"
href="usage-paths/ta-guide.html"
title="next page">
<div class="prev-next-info">
<p class="prev-next-subtitle">next</p>
<p class="prev-next-title">Teaching Assistant Guide for TinyTorch</p>
</div>
<i class="fa-solid fa-angle-right"></i>
</a>
</div>
</footer>
</div>
<div class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
<div class="sidebar-secondary-item">
<div class="page-toc tocsection onthispage">
<i class="fa-solid fa-list"></i> Contents
</div>
<nav class="bd-toc-nav page-toc">
<ul class="visible nav section-nav flex-column">
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#course-overview">🎯 Course Overview</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#instructor-setup">🛠️ Instructor Setup</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#initial-setup"><strong>1. Initial Setup</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#verify-installation"><strong>2. Verify Installation</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#assignment-workflow">📝 Assignment Workflow</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#simplified-with-tito-cli"><strong>Simplified with Tito CLI</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#prepare-assignments"><strong>1. Prepare Assignments</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#distribute-to-students"><strong>2. Distribute to Students</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#collect-submissions"><strong>3. Collect Submissions</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#auto-grade"><strong>4. Auto-Grade</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#manual-review"><strong>5. Manual Review</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#generate-feedback"><strong>6. Generate Feedback</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#export-grades"><strong>7. Export Grades</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#grading-components">📊 Grading Components</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#auto-graded-70"><strong>Auto-Graded (70%)</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#manually-graded-30"><strong>Manually Graded (30%)</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#grading-rubric-for-ml-systems-questions"><strong>Grading Rubric for ML Systems Questions</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#sample-solutions-for-grading-calibration">📋 Sample Solutions for Grading Calibration</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-01-tensor-memory-footprint">Module 01: Tensor - Memory Footprint</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-05-autograd-backward-pass">Module 05: Autograd - Backward Pass</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-09-spatial-convolution-implementation">Module 09: Spatial - Convolution Implementation</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-12-attention-scaled-dot-product-attention">Module 12: Attention - Scaled Dot-Product Attention</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#grading-guidelines-using-sample-solutions">Grading Guidelines Using Sample Solutions</a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-teaching-notes">📚 Module Teaching Notes</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-01-tensor"><strong>Module 01: Tensor</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-02-activations"><strong>Module 02: Activations</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-04-05-layers-networks"><strong>Module 04-05: Layers &amp; Networks</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-06-07-spatial-attention"><strong>Module 06-07: Spatial &amp; Attention</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-08-11-training-pipeline"><strong>Module 08-11: Training Pipeline</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-12-15-production"><strong>Module 12-15: Production</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#module-16-tinygpt"><strong>Module 16: TinyGPT</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#learning-objectives">🎯 Learning Objectives</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#tracking-progress">📈 Tracking Progress</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#individual-progress"><strong>Individual Progress</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#class-overview"><strong>Class Overview</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#identify-struggling-students"><strong>Identify Struggling Students</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#teaching-tips">💡 Teaching Tips</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#emphasize-building-over-theory"><strong>1. Emphasize Building Over Theory</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#connect-to-production-systems"><strong>2. Connect to Production Systems</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#make-performance-visible"><strong>3. Make Performance Visible</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#encourage-systems-questions"><strong>4. Encourage Systems Questions</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#troubleshooting">🔧 Troubleshooting</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#common-student-issues"><strong>Common Student Issues</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#nbgrader-issues"><strong>NBGrader Issues</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#sample-schedule-16-weeks">📊 Sample Schedule (16 Weeks)</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#assessment-strategy">🎓 Assessment Strategy</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#continuous-assessment-70"><strong>Continuous Assessment (70%)</strong></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#projects-30"><strong>Projects (30%)</strong></a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#additional-resources">📚 Additional Resources</a></li>
</ul>
</nav></div>
</div></div>
</div>
<footer class="bd-footer-content">
<div class="bd-footer-content__inner container">
<div class="footer-item">
<p class="component-author">
By Prof. Vijay Janapa Reddi (Harvard University)
</p>
</div>
<div class="footer-item">
<p class="copyright">
© Copyright 2025.
<br/>
</p>
</div>
<div class="footer-item">
</div>
<div class="footer-item">
</div>
</div>
</footer>
</main>
</div>
</div>
<!-- Scripts loaded after <body> so the DOM is not blocked -->
<script src="_static/scripts/bootstrap.js?digest=dfe6caa3a7d634c4db9b"></script>
<script src="_static/scripts/pydata-sphinx-theme.js?digest=dfe6caa3a7d634c4db9b"></script>
<footer class="bd-footer">
</footer>
</body>
</html>
Reference in New Issue View Git Blame Copy Permalink