demo.git/src/tests/gpu, branch main

style: apply clang-format

2026-05-21T06:10:47Z

feat(cnn_v3): upgrade architecture to enc_channels=[8,16]

2026-03-26T06:03:01Z

Double encoder capacity: enc0 4→8ch, enc1 8→16ch, bottleneck 16→16ch, dec1 32→8ch, dec0 16→4ch. Total weights 2476→7828 f16 (~15.3 KB). FiLM MLP output 40→72 params (L1: 16×40→16×72). 16-ch textures split into _lo/_hi rgba32uint pairs (enc1, bottleneck). enc0 and dec1 textures changed from rgba16float to rgba32uint (8ch). GBUF_RGBA32UINT node gains CopySrc for parity test readback. - WGSL shaders: all 5 passes rewritten for new channel counts - C++ CNNv3Effect: new weight offsets/sizes, 8ch uniform structs - Web tool (shaders.js + tester.js): matching texture formats and bindings - Parity test: readback_rgba32uint_8ch helper, updated vector counts - Training scripts: default enc_channels=[8,16], updated docstrings - Docs + architecture PNG regenerated handoff(Gemini): CNN v3 [8,16] upgrade complete. All code, tests, web tool, training scripts, and docs updated. Next: run training pass.

feat(cnn_v3): 3×3 dilated bottleneck + Sobel loss + FiLM warmup + architecture PNG

2026-03-25T09:05:42Z

- Replace 1×1 pointwise bottleneck with Conv(8→8, 3×3, dilation=2): effective RF grows from ~13px to ~29px at ¼res (~+1 KB weights) - Add Sobel edge loss in training (--edge-loss-weight, default 0.1) - Add FiLM 2-phase training: freeze MLP for warmup epochs then unfreeze at lr×0.1 (--film-warmup-epochs, default 50) - Update weight layout: BN 72→584 f16, total 1964→2476 f16 (4952 B) - Cascade offsets in C++ effect, JS tool, export/gen_test_vectors scripts - Regenerate test_vectors.h (1238 u32); parity max_err=9.77e-04 - Generate dark-theme U-Net+FiLM architecture PNG (gen_architecture_png.py) - Replace ASCII art in CNN_V3.md and HOW_TO_CNN.md with PNG embed handoff(Gemini): bottleneck dilation + Sobel loss + FiLM warmup landed. Next: run first real training pass (see cnn_v3/docs/HOWTO.md §3).

feat(gbuffer): wire_dag() + find_downstream_output() for temporal feedback

2026-03-23T06:54:18Z

- Add Effect::wire_dag() virtual (called from init_effect_nodes after full DAG built) - Add Effect::find_downstream_output() protected helper (first downstream consumer output) - GBufferEffect::wire_dag() auto-sets cnn_output_node_ via find_downstream_output, guarding against sink (external view, null texture) - GBufferEffect::post_render() null-checks src texture before CopyTextureToTexture - Tests: find_downstream_output cases + wire_dag integration in test_effect_base - Doc: SEQUENCE.md updated with wire_dag pattern, helper contract, and sink guard Co-Authored-By: Claude Sonnet 4.6

feat(cnn_v3): GBufDeferredEffect — simple deferred render (albedo * shadow)

2026-03-22T18:58:04Z

New effect unpacks feat_tex0/feat_tex1 and outputs albedo * shadow. Replaces CNNv3Effect in cnn_v3_test sequence until training is complete. 37/37 tests passing. handoff(Gemini): GBufDeferredEffect wired in timeline; CNN v3 pipeline: GBufferEffect → GBufDeferredEffect → sink.

feat(cnn_v3): add G-buffer visualizer + web sample loader (Phase 7)

2026-03-22T15:21:25Z

C++ GBufViewEffect: renders all 20 feature channels from feat_tex0/feat_tex1 in a 4×5 tiled grid. Custom BGL with WGPUTextureSampleType_Uint; bind group rebuilt per frame via wgpuRenderPipelineGetBindGroupLayout. Web tool: "Load sample directory" button — webkitdirectory picker, FULL_PACK_SHADER compute (matches gbuf_pack.wgsl packing), runFromFeat() skips photo-pack step, computePSNR() readback + comparison vs target.png side-by-side. 36/36 tests pass. Docs updated: HOWTO.md §9, README, PROJECT_CONTEXT, TODO, COMPLETED. handoff(Gemini): CNN v3 Phase 7 done. Next: run train_cnn_v3.py (see HOWTO §3).

feat(cnn_v3): wire trained weights into CNNv3Effect + add timeline test sequence

2026-03-22T11:53:13Z

- CNNv3Effect constructor loads ASSET_WEIGHTS_CNN_V3 via GetAsset on startup - seq_compiler.py: CLASS_TO_HEADER supports full #include paths for cnn_v3/ classes - timeline.seq: add cnn_v3_test sequence at 48s (GBufferEffect → CNNv3Effect) - test_cnn_v3_parity: zero_weights test now explicitly uploads zeros to override asset handoff(Gemini): CNNv3Effect ready; export weights to workspaces/main/weights/ and seek to 48s to test

refactor(cnn_v3): code review — comments, simplifications, test fix

2026-03-21T13:01:30Z

C++: - cnn_v3_effect.cc: fix declare_nodes comment (output node declared by caller) - cnn_v3_effect.cc: add TODO(phase-7) marker for FiLM MLP replacement WGSL: - cnn_v3_bottleneck.wgsl: consolidate _pad fields onto one line, explain why array is invalid in uniform address space - cnn_v3_enc0.wgsl: fix "12xu8" → "12ch u8norm" in header comment - cnn_v3_dec0.wgsl: clarify parity note (sigmoid after FiLM+ReLU, not raw conv) - cnn_v3_common.wgsl: clarify unpack_8ch pack layout (low/high 16 bits) Python: - cnn_v3_utils.py: replace PIL-based _upsample_nearest (uint8 round-trip) with pure numpy index arithmetic; rename _resize_rgb → _resize_img (handles any channel count); add comment on normal zero-pad workaround - export_cnn_v3_weights.py: add cross-ref to cnn_v3_effect.cc constants; clarify weight count comments with Conv notation Test: - test_cnn_v3_parity.cc: enc0/dec1 layer failures now return 0 (were print-only) handoff(Gemini): CNN v3 review complete, 36/36 tests passing.

feat(cnn_v3): Phase 5 complete — parity validation passing (36/36 tests)

2026-03-21T08:51:58Z

- Add test_cnn_v3_parity.cc: zero_weights + random_weights tests - Add gen_test_vectors.py: PyTorch reference implementation for enc0/enc1/bn/dec1/dec0 - Add test_vectors.h: generated C header with enc0, dec1, output expected values - Fix declare_nodes(): intermediate textures at fractional resolutions (W/2, W/4) using new NodeRegistry::default_width()/default_height() getters - Add layer-by-layer readback (enc0, dec1) for regression coverage - Final parity: enc0 max_err=1.95e-3, dec1 max_err=1.95e-3, out max_err=4.88e-4 handoff(Claude): CNN v3 parity done. Next: train_cnn_v3.py (FiLM MLP training).

feat(cnn_v3): Phase 4 complete — CNNv3Effect C++ + FiLM uniform upload

2026-03-21T07:52:53Z

- cnn_v3/src/cnn_v3_effect.{h,cc}: full Effect subclass with 5 compute passes (enc0→enc1→bottleneck→dec1→dec0), shared weights storage buffer, per-pass uniform buffers, set_film_params() API - Fixed WGSL/C++ struct alignment: vec3u has align=16, so CnnV3Params4ch is 64 bytes and CnnV3ParamsEnc1 is 96 bytes (not 48/80) - Weight offsets computed as explicit formulas (e.g. 20*4*9+4) for clarity - Registered in CMake, shaders.h/cc, demo_effects.h, test_demo_effects.cc - 35/35 tests pass handoff(Gemini): CNN v3 Phase 5 next — parity validation (Python ref vs WGSL)