Skip to content

Conversation

@Lunderberg
Copy link
Contributor

The relax::op::linear is implemented as permute_dims, followed by matmul. In this case, readability can be improved by naming the weights. For example, if the weights provided to nn.Linear are named linear_layer_1_weight, then the weights with permuted dimensions can be named matmul_layer_1_weight instead of permute_dims.

The `relax::op::linear` is implemented as `permute_dims`, followed by
`matmul`.  In this case, readability can be improved by naming the
weights.
Copy link
Member

@junrushao junrushao left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM!

@junrushao junrushao merged commit f215a41 into apache:unity Jan 5, 2024
@Lunderberg Lunderberg deleted the unity_nn_linear_name branch January 5, 2024 18:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants