Fix CI: Add type assertions for TTQ layer parameters

- Add runtime type assertions in TTQ forward() methods
- Resolves mypy errors about Parameter | Tensor types
- Parameters are always Tensors after register_parameter()

Author: dariocazzani

bitnet/nn/ttq_conv2d.py

@@ -40,5 +40,10 @@ def inv_softplus(x: float) -> float:
     def forward(self, x: Tensor) -> Tensor:
         # Pure TTQ: Only quantize weights, use FP32 activations
         # This is TTQ as described in the original paper
+        # Type assertions for mypy - these are always Tensors after register_parameter
+        assert isinstance(self.wp, Tensor)
+        assert isinstance(self.wn, Tensor)
+        assert isinstance(self.delta, Tensor)
+
         w_quant, _, _ = ttq_quantize(self.weight, self.wp, self.wn, self.delta)
         return f.conv2d(x, w_quant, self.bias, self.stride, self.padding, self.dilation, self.groups)

bitnet/nn/ttq_linear.py

@@ -45,5 +45,10 @@ def inv_softplus(x: float) -> float:
     def forward(self, x: Tensor) -> Tensor:
         # Pure TTQ: Only quantize weights, use FP32 activations
         # This is TTQ as described in the original paper
+        # Type assertions for mypy - these are always Tensors after register_parameter
+        assert isinstance(self.wp, Tensor)
+        assert isinstance(self.wn, Tensor)
+        assert isinstance(self.delta, Tensor)
+
         w_quant, _, _ = ttq_quantize(self.weight, self.wp, self.wn, self.delta)
         return f.linear(x, w_quant, self.bias)