Rename from tiny-mod12-verified

README.md

@@ -0,0 +1,72 @@
+---
+license: mit
+tags:
+- pytorch
+- safetensors
+- threshold-logic
+- neuromorphic
+- modular-arithmetic
+---
+
+# threshold-mod12
+
+Trivial case: computes Hamming weight mod 12 for 8-bit inputs. Since max HW is 8 < 12, this is just HW.
+
+## Circuit
+
+```
+  x₀ x₁ x₂ x₃ x₄ x₅ x₆ x₇
+   │  │  │  │  │  │  │  │
+   │  │  │  │  │  │  │  │
+   w: 1  1  1  1  1  1  1  1
+   └──┴──┴──┴──┼──┴──┴──┴──┘
+               ▼
+          ┌─────────┐
+          │ b:  0   │
+          └─────────┘
+               │
+               ▼
+       HW (= HW mod 12)
+```
+
+## Why Trivial?
+
+For mod m where m > (number of inputs), no reset ever occurs:
+
+- 8 inputs → max HW = 8
+- 8 mod 12 = 8 (no wraparound)
+
+## Parameters
+
+| | |
+|---|---|
+| Weights | [1, 1, 1, 1, 1, 1, 1, 1] |
+| Bias | 0 |
+| Total | 9 parameters |
+
+## Usage
+
+```python
+from safetensors.torch import load_file
+import torch
+
+w = load_file('model.safetensors')
+
+def mod12(bits):  # Actually just HW
+    inputs = torch.tensor([float(b) for b in bits])
+    return int((inputs * w['weight']).sum() + w['bias'])
+```
+
+## Files
+
+```
+threshold-mod12/
+├── model.safetensors
+├── model.py
+├── config.json
+└── README.md
+```
+
+## License
+
+MIT

config.json

@@ -0,0 +1,23 @@
+{
+  "model_type": "threshold_network",
+  "task": "mod12_classification",
+  "architecture": "8 -> 1",
+  "input_size": 8,
+  "output_size": 1,
+  "num_neurons": 1,
+  "num_parameters": 9,
+  "modulus": 12,
+  "activation": "heaviside",
+  "weight_constraints": "integer",
+  "weight_pattern": "[1, 1, 1, 1, 1, 1, 1, 1]",
+  "verification": {
+    "method": "coq_proof",
+    "exhaustive": true,
+    "inputs_tested": 256
+  },
+  "accuracy": {
+    "all_inputs": "256/256",
+    "percentage": 100.0
+  },
+  "github": "https://github.com/CharlesCNorton/coq-circuits"
+}

model.py

@@ -0,0 +1,22 @@
+"""
+Threshold Network for MOD-12 Circuit
+
+For 8-bit inputs, HW ranges 0-8, all less than 12, so HW mod 12 = HW.
+"""
+
+import torch
+from safetensors.torch import load_file
+
+
+class ThresholdMod12:
+    def __init__(self, weights_dict):
+        self.weight = weights_dict['weight']
+        self.bias = weights_dict['bias']
+
+    def __call__(self, bits):
+        inputs = torch.tensor([float(b) for b in bits])
+        return (inputs * self.weight).sum() + self.bias
+
+    @classmethod
+    def from_safetensors(cls, path="model.safetensors"):
+        return cls(load_file(path))

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d9838dc724da91cd61df87cf34f8fc697e97d3f2b257ceceb971995deed4643
+size 164