[README] - update faces_sub_image entry; bump N_HID to 256

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

src/tests/README.md

@@ -29,7 +29,7 @@ These load images from a fixed path on disk.
 
 | File | RBM type | What it tests |
 |---|---|---|
-| `test_faces_sub_image.py` | GB-RBM | Auto-encoder on 32×32 RGB patches extracted from Caltech WebFaces (`/media/jens/cifs/bilder/MachineVision/Caltech_WebFaces/`). Uses `SubImage` for non-overlapping ROI extraction (stride=32). After training: shows 128 learned weight filters, patch-level reconstructions, and a full image reconstruction assembled from patches. |
+| `test_faces_sub_image.py` | GB-RBM | Auto-encoder on 32×32 patches from Caltech WebFaces (`/media/jens/cifs/bilder/MachineVision/Caltech_WebFaces/`). Uses `SubImage` with 50% overlap (stride=16); reconstruction blends overlapping patches by pixel-averaging. Supports RGB and grayscale modes. CLI args: `--grayscale`, `--load_model`, `--do_train`, `--l1_lambda`. After training: shows learned weight filters, patch-level reconstructions, and a full overlap-blended image reconstruction. Default: 256 hidden units, 1000 epochs, mini_batch=1000. **L1=0.01 is the recommended sweet spot** — produces structured edge/texture filters with 15% sparsity while keeping MSE~0.242. L1=0.05 collapses most filters to blob detectors (58% sparsity, MSE~0.261). |
 | `test_sub_image.py` | — | Unit test for `SubImage` (patch extraction). No RBM involved. |
 | `test_conv2d.py` | — | Unit test for `conv2d`. No RBM involved. |
 

src/tests/test_faces_sub_image.py

@@ -14,7 +14,7 @@ STRIDE    = 16    # 50% overlap; set equal to PATCH for non-overlapping
 GRAYSCALE = False  # reassigned in __main__ when --grayscale is set
 N_CH      = 1 if GRAYSCALE else 3
 N_VIS     = N_CH * PATCH * PATCH   # 1024 grayscale / 3072 colour
-N_HID     = 128
+N_HID     = 256
 N_IMAGES  = 50