Skip to content

Commit

Permalink
Fix (ptq): disable MHA input_quant in graph quant
Browse files Browse the repository at this point in the history
  • Loading branch information
Giuseppe5 committed Nov 28, 2023
1 parent 63370f4 commit 4aa9968
Showing 1 changed file with 1 addition and 0 deletions.
Original file line number Diff line number Diff line change
Expand Up @@ -345,6 +345,7 @@ def kwargs_prefix(prefix, weight_kwargs):
quant_mha_kwargs = {
**kwargs_prefix('in_proj_', weight_quant_dict),
**kwargs_prefix('out_proj_', weight_quant_dict),
'in_proj_input_quant': None,
'in_proj_bias_quant': bias_quant,
'softmax_input_quant': None,
'attn_output_weights_quant': sym_act_quant,
Expand Down

0 comments on commit 4aa9968

Please sign in to comment.