Skip to content

Commit 12e4fc1

Browse files
julesmuhizijmduarte
authored andcommitted
update utils.py for support QDenseBN quant model save
1 parent 476e371 commit 12e4fc1

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

qkeras/utils.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -383,7 +383,7 @@ def model_save_quantized_weights(model, filename=None):
383383
if has_scale:
384384
saved_weights[layer.name]["scales"] = scales
385385
if not any(isinstance(layer, t) for t in [
386-
QConv2DBatchnorm, QDepthwiseConv2DBatchnorm]):
386+
QConv2DBatchnorm, QDenseBatchnorm, QDepthwiseConv2DBatchnorm]):
387387
# Set layer weights in the format that software inference uses
388388
layer.set_weights(weights)
389389
else:

0 commit comments

Comments
 (0)