Removed:
- Detailed submission infrastructure CLI commands
- Adoption tracking metrics
- Promotional language about leaderboards
Kept:
- MLSysBook ecosystem integration
- Pedagogical value of competitive benchmarking (Module 20)
- Focus on systems thinking and measurement-driven decisions
The section now focuses on educational value rather than infrastructure details.
Paper additions based on student feedback:
- MLSysBook ecosystem integration in Architecture section
- Hardware simulation integration (scale-sim, timeloop, astra-sim) in Future Work
- Enhanced community sustainability discussion
- Bibliography entries for MLSysBook textbook and hardware simulators
Addresses feedback from Zishen Wan on:
- Connecting TinyTorch to broader ML Systems Book curriculum
- System simulator integration for hardware performance analysis
- Community infrastructure and sustainability
- Clarifies this is the Harvard/MLSysBook TinyTorch
- Acknowledges 5+ other TinyTorch educational implementations
- Highlights unique features: 20-module curriculum, NBGrader, systems focus
- Links to ML Systems Book ecosystem (mlsysbook.ai/tinytorch)
- Community-positive framing
Removed from src/:
- 4 .ipynb files (auto-generated, belong in modules/)
- autograd_systems_analysis.py (supplementary content without export directives)
- validate_fixes.py (temporary validation script)
Source directory now contains only:
- One .py file per module (01_tensor.py through 20_capstone.py)
- ABOUT.md files (module documentation)
- No temporary or auto-generated files
This ensures src/ is the clean source of truth for all 20 modules.
Refines diagrams across multiple modules to enhance
readability and maintain a consistent visual style.
This change improves the overall clarity and understandability
of the documentation and explanations within the codebase.
- Update paper/paper.tex to reflect Module 20 submission infrastructure
- Add nbdev export integration to paper build system section
- Integrate community submission workflow into paper
- Enhance Module 20 with ~4,500 words of pedagogical content
- Add 15+ ASCII diagrams for visual learning
- Include comprehensive benchmarking foundations
- Add module summary celebrating 20-module journey
- Complete pre-release review (96/100 - ready for release)
Implement tito community submit command for Module 20 capstone submissions with schema validation, metrics display, and error handling. Shows baseline and optimized model metrics with improvement calculations. Includes Coming soon message for future leaderboard integration.
Module 20 now demonstrates the complete benchmarking workflow:
- SimpleMLP toy model for demonstration (no milestone dependencies)
- BenchmarkReport class for measuring performance metrics
- generate_submission() function for creating JSON submissions
- Complete example workflow students can modify
- All tests pass
This launch-ready module shows students how to:
1. Benchmark a model using Module 19 tools
2. Generate standardized JSON submissions
3. Share results with the TinyTorch community
Exports to: tinytorch.capstone
Restored the original competition-focused Module 20 from git history.
The previous TinyGPT-focused version was replaced with the intended
competition and submission generation module.
Original Module 20 purpose:
- TinyTorch Olympics competition framework
- Uses benchmarking harness from Module 19
- Generates MLPerf-style JSON submissions
- Olympic events: Latency Sprint, Memory Challenge, Accuracy Contest, etc.
- Exports to tinytorch.competition.submit
Fixed imports to match current Module 19:
- Changed from BenchmarkResult to Benchmark, BenchmarkSuite, TinyMLPerf
- Added missing time import
Note: Module still needs additional fixes to pass tests (validation logic).
This commit restores the correct architectural direction for Module 20.
Module 20 (Capstone) had an unused matplotlib.pyplot import that was
causing tests to fail when matplotlib wasn't installed.
The import was a leftover from early development but matplotlib is
never actually used in the module (no plt.* calls anywhere).
Module 20 is a capstone integration module that:
- Imports and integrates all 19 previous TinyTorch modules
- Exports TinyGPT, TinyGPTTrainer, and CompleteTinyGPTPipeline
- Demonstrates the complete framework working together
- Should have zero external dependencies beyond numpy
Removing this dependency ensures Module 20 can run in minimal
environments with only numpy and the TinyTorch modules.
Module 09's main block was calling analyze_convolution_complexity() and
analyze_pooling_effects() before test_module(). These analysis functions
are educational demonstrations that:
- Run computational benchmarks with timing
- Test multiple configurations for performance analysis
- Take significant time to execute
During 'tito module test', we only want to run test_module() to verify
correctness, not run performance benchmarks. This reduces Module 09
test time significantly (from ~30+ seconds to ~12 seconds).
Analysis functions remain in the module for educational purposes but
are not exported and not called during standard testing.
All other modules (01-20) already follow this pattern correctly.
Added live progress bar to `tito module test --all` command:
- Shows spinner, progress bar, and task progress (X/20)
- Updates description with current module being tested
- Provides better visual feedback during long test runs
- Maintains all existing rich output (logo, panels, tables)
Uses rich.progress with SpinnerColumn, TextColumn, BarColumn,
and TaskProgressColumn for comprehensive progress visualization.
This completes the rich CLI enhancement pass for bulk commands.
New Commands:
- tito module test [NUMBER] - Test single module
- tito module test --all - Test all 20 modules sequentially
- tito module complete --all - Complete all modules (test + export)
- tito module reset --all - Reset all modules (already existed)
Features:
- Detailed test results with pass/fail status
- --verbose flag for full test output
- --stop-on-fail to halt on first failure
- Summary table showing all module test results
- 5-minute timeout per module test
- Proper error reporting and exit codes
This enables:
- Quick validation of all modules after global changes
- Bulk export workflow for package releases
- Easy testing during development
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Updated all carousel demo GIFs using new tape scripts:
- 00-welcome.gif: Updated with latest tape
- 02-build-test-ship.gif: Regenerated with clean setup phase
- 03-milestone-unlocked.gif: Regenerated with improved flow
- 04-share-journey.gif: Regenerated with better presentation
All demos now show cleaner, more professional output.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Demo improvements:
- Add hidden setup phase to demo tapes for clean state
- New benchmark and logo demo tapes
- Improved build-test-ship, milestone, and share-journey demos
- All demos now use Hide/Show for cleaner presentation
CLI fix:
- Add default=None to module reset command argument
- Prevents argparse error when no module specified
Cleanup:
- Remove outdated tinytorch/core/activations.py binary
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Changed closing from "Let's build ML from scratch!" to
"Let's recreate PyTorch together!" for clarity and
better framing of TinyTorch's goal.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Fixed broken multiline comment that had closing parenthesis on wrong line.
The syntax error was preventing tito from running.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Enhanced CIFAR-10 CNN with BatchNorm2d for stable training
- Added RandomHorizontalFlip and RandomCrop augmentation transforms
- Improved training accuracy from 65%+ to 70%+ with modern architecture
- Updated demo tapes with opening comments for clarity
- Regenerated welcome GIF, removed outdated demo GIFs
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Typography refinements:
- Add extra space between TORCH letters for better readability
- Improve breathing room in R, C, and H letter forms
- Adjust flame positioning to match wider letter spacing
- Align tagline with updated TORCH width
The wider spacing gives each letter more presence and makes
the logo feel less cramped on narrow terminal windows.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Typography refinements:
- Move flames 1 space closer to T and H for better visual cohesion
- Make tagline bold to match flame glow intensity
- Use TAGLINE_COLOR constant for consistency
These micro-adjustments make the logo feel crafted rather than
assembled, with each element belonging to a unified whole.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
All demos now:
- Start with opening comment explaining what will be shown
- Show cd and source activate.sh commands to users
- Use custom TinyTorch theme colors
- Only hide fast-forward module completions (Demo 03)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Added fourth section with essential help commands:
- tito system doctor (check environment health)
- tito --help (see all commands)
Completes welcome screen simplification with 4 focused groups
showing 10 total commands for new users.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Added proper spacing (8 spaces) to align the tagline
'🔥 Don't just import it. Build it.' directly under the TORCH letters.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Key changes:
- SHOW cd and source activate.sh in all demos (users see full setup)
- ONLY use Hide for Demo 03's module 01-06 completions (fast-forward)
- Remove unnecessary clear command
- Add module reset to demo script for clean slate
This ensures users see the real workflow while keeping demos concise.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
