Update int_torch_handler.py

Xilinx · Oct 30, 2024 · 60367a6 · 60367a6
1 parent dd05b52
commit 60367a6
Showing 1 changed file with 0 additions and 2 deletions.
diff --git a/src/brevitas/quant_tensor/int_torch_handler.py b/src/brevitas/quant_tensor/int_torch_handler.py
@@ -234,8 +234,6 @@ def quant_output_scale_impl(
     output_scale_shape = compute_channel_view_shape(inp, channel_dim=channel_dim)
 
     quant_weight_scale = quant_weight_scale.view(output_scale_shape)
-    if len(quant_input_scale.shape) == 0:
-        quant_input_scale = quant_input_scale.view(output_scale_shape)
     quant_input_scale = quant_input_scale.view(output_scale_shape)
     if not is_broadcastable(quant_weight_scale.shape, quant_input_scale.shape):
         return None