5-input minority gate, magnitude 7

README.md

@@ -0,0 +1,53 @@
+---
+license: mit
+tags:
+- pytorch
+- safetensors
+- threshold-logic
+- neuromorphic
+---
+
+# threshold-minority5
+
+5-input minority gate. Outputs 1 when at most 2 of 5 inputs are high. Complement of majority5.
+
+## Function
+
+minority5(a, b, c, d, e) = 1 if sum <= 2, else 0
+
+## Architecture
+
+Single neuron: weights [-1, -1, -1, -1, -1], bias 2
+
+Fires when: sum <= 2
+
+## Parameters
+
+| | |
+|---|---|
+| Inputs | 5 |
+| Outputs | 1 |
+| Neurons | 1 |
+| Layers | 1 |
+| Parameters | 6 |
+| Magnitude | 7 |
+
+## Usage
+
+```python
+from safetensors.torch import load_file
+import torch
+
+w = load_file('model.safetensors')
+
+def minority5(a, b, c, d, e):
+    inp = torch.tensor([float(a), float(b), float(c), float(d), float(e)])
+    return int((inp @ w['neuron.weight'].T + w['neuron.bias'] >= 0).item())
+
+print(minority5(0, 0, 0, 1, 1))  # 1
+print(minority5(0, 0, 1, 1, 1))  # 0
+```
+
+## License
+
+MIT

config.json

@@ -0,0 +1,9 @@
+{
+  "name": "threshold-minority5",
+  "description": "5-input minority gate",
+  "inputs": 5,
+  "outputs": 1,
+  "neurons": 1,
+  "layers": 1,
+  "parameters": 6
+}

create_safetensors.py

@@ -0,0 +1,24 @@
+import torch
+from safetensors.torch import save_file
+
+weights = {
+    'neuron.weight': torch.tensor([[-1.0, -1.0, -1.0, -1.0, -1.0]], dtype=torch.float32),
+    'neuron.bias': torch.tensor([2.0], dtype=torch.float32)
+}
+save_file(weights, 'model.safetensors')
+
+def minority5(*args):
+    inp = torch.tensor([float(x) for x in args])
+    return int((inp @ weights['neuron.weight'].T + weights['neuron.bias'] >= 0).item())
+
+print("Verifying minority5...")
+errors = 0
+for i in range(32):
+    bits = [(i >> j) & 1 for j in range(4, -1, -1)]
+    result = minority5(*bits)
+    expected = 1 if sum(bits) <= 2 else 0
+    if result != expected:
+        errors += 1
+if errors == 0:
+    print("All 32 test cases passed!")
+print(f"Magnitude: {sum(t.abs().sum().item() for t in weights.values()):.0f}")

model.py

@@ -0,0 +1,16 @@
+import torch
+from safetensors.torch import load_file
+
+def load_model(path='model.safetensors'):
+    return load_file(path)
+
+def minority5(a, b, c, d, e, weights):
+    inp = torch.tensor([float(a), float(b), float(c), float(d), float(e)])
+    return int((inp @ weights['neuron.weight'].T + weights['neuron.bias'] >= 0).item())
+
+if __name__ == '__main__':
+    w = load_model()
+    print('minority5 sample outputs:')
+    print(f'  00000 -> {minority5(0,0,0,0,0,w)}')
+    print(f'  00011 -> {minority5(0,0,0,1,1,w)}')
+    print(f'  00111 -> {minority5(0,0,1,1,1,w)}')