HDCharles 428cbd7513 [ao] fixing multihead attention convert size (#110407) ... Summary: after converting nn.multihead attention we weren't deleting the old in_proj_weight and in_proj_bias despite not (really) using them. Test Plan: python test/test_quantization.py -k "test_custom_module_multi_head_attention" Reviewers: Subscribers: Tasks: Tags: Pull Request resolved: https://github.com/pytorch/pytorch/pull/110407 Approved by: https://github.com/jerryzh168		2023-10-03 08:49:12 +00:00
..
__init__.py	[BE] Enable ruff's UP rules and autoformat ao/ (#105430)	2023-07-19 13:44:37 +00:00
activation.py	[ao] fixing multihead attention convert size (#110407)	2023-10-03 08:49:12 +00:00
batchnorm.py
conv.py	[BE] Enable ruff's UP rules and autoformat ao/ (#105430)	2023-07-19 13:44:37 +00:00
dropout.py
embedding_ops.py
functional_modules.py	[pytorch][ao] Add torch.matmul in FloatFunctional/QFunctional (#106831)	2023-08-10 22:43:36 +00:00
linear.py	[BE] Enable ruff's UP rules and autoformat ao/ (#105430)	2023-07-19 13:44:37 +00:00
normalization.py
rnn.py
utils.py