diff options
| author | skal <pascal.massimino@gmail.com> | 2026-02-11 23:35:44 +0100 |
|---|---|---|
| committer | skal <pascal.massimino@gmail.com> | 2026-02-11 23:35:44 +0100 |
| commit | 409bbfb08fae03bfb7daa554a799bd8480806799 (patch) | |
| tree | 551e02036b58915718b6eb1ad127f19766d21633 /workspaces/test/shaders | |
| parent | 75b820e1d5be15b0187bb201ca432157b4049bc5 (diff) | |
docs: Add CNN flatten mode technical analysis
Comprehensive analysis of single-pass CNN shader architecture:
- Full flatten (3 layers): 544 bytes/thread register pressure - NOT recommended
- Partial flatten (layers 1+2): 288 bytes/thread - marginal benefit
- Current multi-pass: Optimal for GPU occupancy and maintainability
Recommendation: Keep current 3-pass architecture.
Alternative size optimizations: weight quantization, kernel reduction.
handoff(Claude): CNN flatten analysis documented
Diffstat (limited to 'workspaces/test/shaders')
0 files changed, 0 insertions, 0 deletions
