CNN v2: Alpha channel depth handling and layer visualization

Training changes:
- Changed p3 default depth from 0.0 to 1.0 (far plane semantics)
- Extract depth from target alpha channel in both datasets
- Consistent alpha-as-depth across training/validation

Test tool enhancements (cnn_test):
- Added load_depth_from_alpha() for R32Float depth texture
- Fixed bind group layout for UnfilterableFloat sampling
- Added --save-intermediates with per-channel grayscale composites
- Each layer saved as 4x wide PNG (p0-p3 stacked horizontally)
- Global layers_composite.png for vertical layer stack overview

Investigation notes:
- Static features p4-p7 ARE computed and bound correctly
- Sin_20_y pattern visibility difference between tools under investigation
- Binary weights timestamp (Feb 13 20:36) vs HTML tool (Feb 13 22:12)
- Next: Update HTML tool with canonical binary weights

handoff(Claude): HTML tool weights update pending - base64 encoded
canonical weights ready in /tmp/weights_b64.txt for line 392 replacement.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

1 files changed, 0 insertions, 43 deletions

diff --git a/LOG.txt b/LOG.txt
deleted file mode 100644
index 50b77ea..0000000
--- a/LOG.txt
+++ /dev/null
@@ -1,43 +0,0 @@
-=== CNN v2 Complete Training Pipeline ===
-Input: training/input
-Target: training/target_2
-Epochs: 10000
-Checkpoint interval: 500
-
-[1/4] Training CNN v2 model...
-Training on cpu
-Loaded 8 image pairs
-Model: [16, 8, 4] channels, [1, 3, 5] kernels, 3456 weights
-
-Training for 10000 epochs...
-Traceback (most recent call last):
-  File "/Users/skal/demo/training/train_cnn_v2.py", line 217, in <module>
-    main()
-  File "/Users/skal/demo/training/train_cnn_v2.py", line 213, in main
-    train(args)
-  File "/Users/skal/demo/training/train_cnn_v2.py", line 157, in train
-    for static_feat, target in dataloader:
-  File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/torch/utils/data/dataloader.py", line 741, in __next__
-    data = self._next_data()
-           ^^^^^^^^^^^^^^^^^
-  File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/torch/utils/data/dataloader.py", line 801, in _next_data
-    data = self._dataset_fetcher.fetch(index)  # may raise StopIteration
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/torch/utils/data/_utils/fetch.py", line 57, in fetch
-    return self.collate_fn(data)
-           ^^^^^^^^^^^^^^^^^^^^^
-  File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/torch/utils/data/_utils/collate.py", line 401, in default_collate
-    return collate(batch, collate_fn_map=default_collate_fn_map)
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/torch/utils/data/_utils/collate.py", line 214, in collate
-    return [
-           ^
-  File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/torch/utils/data/_utils/collate.py", line 215, in <listcomp>
-    collate(samples, collate_fn_map=collate_fn_map)
-  File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/torch/utils/data/_utils/collate.py", line 155, in collate
-    return collate_fn_map[elem_type](batch, collate_fn_map=collate_fn_map)
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/torch/utils/data/_utils/collate.py", line 275, in collate_tensor_fn
-    return torch.stack(batch, 0, out=out)
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-RuntimeError: stack expects each tensor to be equal size, but got [8, 376, 626] at entry 0 and [8, 344, 361] at entry 1