File size: 1,434 Bytes
9c4af73 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
"""
Threshold Network for 2-out-of-8 Gate
A formally verified single-neuron threshold network.
Outputs 1 when at least 2 of the 8 inputs are true.
"""
import torch
from safetensors.torch import load_file
class Threshold2OutOf8:
"""
2-out-of-8 threshold gate.
Circuit: output = (sum of inputs - 2 >= 0)
Fires when hamming weight >= 2.
"""
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])
weighted_sum = (inputs * self.weight).sum() + self.bias
return (weighted_sum >= 0).float()
@classmethod
def from_safetensors(cls, path="model.safetensors"):
return cls(load_file(path))
def forward(x, weights):
x = torch.as_tensor(x, dtype=torch.float32)
weighted_sum = (x * weights['weight']).sum(dim=-1) + weights['bias']
return (weighted_sum >= 0).float()
if __name__ == "__main__":
weights = load_file("model.safetensors")
model = Threshold2OutOf8(weights)
print("2-out-of-8 Gate Tests:")
print("-" * 35)
for hw in range(9):
bits = [1]*hw + [0]*(8-hw)
out = int(model(bits).item())
expected = 1 if hw >= 2 else 0
status = "OK" if out == expected else "FAIL"
print(f"HW={hw}: {out} [{status}]")
|