summaryrefslogtreecommitdiff
AgeCommit message (Expand)Author
4 daysopt: Move invariant in1 calculation outside CNN convolution loopsskal
4 daysopt: Vec4-optimize CNN convolution shaders for SIMDskal
4 dayschore: Update CNN architecture to 3×3×3 with new trained weightsskal
4 daysdocs: Update CNN training documentation with patch extractionskal
4 daysfeat: Add salient-point patch extraction for CNN trainingskal
4 daysfix: Use ClampToEdge sampler for CNN to avoid edge wrappingskal
4 daysfix: Correct UV coordinate computation to match PyTorch linspaceskal
4 daysfix: Add clamp to CNN final layer to match PyTorch trainingskal
4 daysfix: PostProcessHelperTest signature and fixture sharingskal
4 daysrefactor: Optimize CNN grayscale computationskal
4 daysupdate train_cnn.py and shaderskal
4 daysdocs: Add inference mode to training documentationskal
4 daysfeat: Add inference mode to train_cnn.py for ground truth generationskal
4 daysfix: CNN training normalization pipeline consistencyskal
4 daysudpate CNN shader code.skal
4 daysfeat: Add --shader-only option to convert_shadertoy.pyskal
4 daysrefactor: Optimize CNN normalization to eliminate redundant conversionsskal
4 daysupdate timeline.seqskal
4 daysfix: Flip Y-axis to match ShaderToy coordinate conventionskal
4 daysrefactor: Improve convert_shadertoy.py to generate compile-ready codeskal
4 daysfeat: Add Scene1 effect from ShaderToy (raymarching cube & sphere)skal
4 dayschore: Remove incomplete CubeSphere effectskal
4 daysdocs: Fix EFFECT keyword syntax and add automation-friendly workflowskal
4 daysfeat: Add ShaderToy conversion toolsskal
4 daysfix: Support variable kernel sizes in CNN layer generationskal
4 daysdocs: Document WGPU builder refactoring in COMPLETED.mdskal
4 daysrefactor: Factor WGPU boilerplate into builder pattern helpersskal
4 daysfeat: CNN RGBD→grayscale with 7-channel augmented inputskal
4 daysudpateskal
4 daysupdate timelineskal
4 daysfix: Resolve CNN effect black screen bug (framebuffer capture + uniforms)skal
4 daysfix: Capture scene framebuffer before post-processing for CNN effectskal
4 daysfeat: Add multi-layer CNN support with framebuffer capture and blend controlskal
4 daysdocs: Update and streamline CNN training documentationskal
4 daysfeat: Add checkpointing support to CNN training scriptskal
4 daysfix: Auto-expand single kernel size to all layers in training scriptskal
4 daysupdate target imagesskal
4 daysfeat: Add coordinate-aware CNN layer 0 for position-dependent stylizationskal
4 daysdocs: Add CNN effect documentation and update project statusskal
4 daysfeat: Add CNN post-processing effect with modular WGSL architectureskal
4 daysinitial doc for the CNN projectskal
5 daysdocs: Reorganize shader editor documentationskal
5 daysfix: Use placeholder technique to prevent syntax highlighting overlapskal
5 daysrefactor: Improve syntax highlighting to avoid overlapping matchesskal
5 daysfeat: Add WGSL syntax highlightingskal
5 daysrefactor: Increase editor pane size by 20% (36% → 43%)skal
5 daysfix: Remove resolutionSelect references after removing UI elementskal
5 daysfeat: Remove resolution selector, enable autoplay on loadskal
5 daysfix: Use 16:9 aspect ratio and object-fit contain for canvasskal
5 daysfix: Canvas fills available height instead of maintaining square aspectskal