Skip to content

Commit 359e9d3

Browse files
authored
Fix custom annotation (#6978)
Summary: Looks like it's added in #6849, maybe it was using the old api for the default 8bit quantization Differential Revision: D66219251
1 parent 84222a9 commit 359e9d3

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

backends/qualcomm/quantizer/custom_annotation.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -78,7 +78,7 @@ def annotate_single_in_single_out(
7878
)
7979

8080
def annotate_matmul_input1(node: Node):
81-
quantization_config_8a8w = get_default_8bit_qnn_ptq_config(
81+
quantization_config_8a8w = get_8a8w_qnn_ptq_config(
8282
act_symmetric=True, act_observer=MinMaxObserver
8383
)
8484
while isinstance(node, Node) and node.op == "call_function":

0 commit comments

Comments
 (0)