Update resnet.py

src/brevitas_examples/bnn_pynq/models/resnet.py

@@ -10,6 +10,7 @@
 from brevitas.quant import Int8WeightPerChannelFloat
 from brevitas.quant import Int8WeightPerTensorFloat
 from brevitas.quant import TruncTo8bit
+from brevitas.quant import IntBias
 from brevitas.quant_tensor import QuantTensor
 
 
@@ -120,6 +121,7 @@ def __init__(
             act_bit_width=8,
             weight_bit_width=8,
             round_average_pool=False,
+            last_layer_bias_quant=IntBias,
             weight_quant=Int8WeightPerChannelFloat,
             first_layer_weight_quant=Int8WeightPerChannelFloat,
             last_layer_weight_quant=Int8WeightPerTensorFloat):
@@ -163,6 +165,7 @@ def __init__(
             num_classes,
             weight_bit_width=8,
             bias=True,
+            bias_quant=last_layer_bias_quant,
             weight_quant=last_layer_weight_quant)
 
         for m in self.modules():
@@ -224,8 +227,9 @@ def quant_resnet18(cfg) -> QuantResNet:
     act_bit_width = cfg.getint('QUANT', 'ACT_BIT_WIDTH')
     num_classes = cfg.getint('MODEL', 'NUM_CLASSES')
     model = QuantResNet(
-        QuantBasicBlock, [2, 2, 2, 2],
+        block_impl=QuantBasicBlock,
+        num_blocks=[2, 2, 2, 2],
         num_classes=num_classes,
         weight_bit_width=weight_bit_width,
         act_bit_width=act_bit_width)
-    return model
+    return model