Skip to content
This repository has been archived by the owner on Oct 13, 2021. It is now read-only.

Handle conv layer spec when input0 is type SpaceToBatchND #475

Merged
merged 50 commits into from
May 4, 2020

Conversation

jiafatom
Copy link
Collaborator

@jiafatom jiafatom commented May 4, 2020

No description provided.

jiafatom and others added 30 commits February 26, 2020 11:38
…nnx#394)

* Efficient-net test cases.

* Add DepthwiseConv2d to subclassed model.

* disable efficient net test cases.
@jiafatom jiafatom requested a review from wenbingl May 4, 2020 21:15
tests/test_layers.py Outdated Show resolved Hide resolved
name=operator.full_name + '_pad_0')
padded_block_prod = oopb.apply_pad(block_prod,
name=operator.full_name + '_pad_0',
pads=[0, 3],
Copy link
Member

Choose a reason for hiding this comment

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

where does pads come from?

Copy link
Collaborator Author

@jiafatom jiafatom May 4, 2020

Choose a reason for hiding this comment

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

This is used for division, it is the dimension of the channel, so 0 means N, 3 means C. It means no division for these two channels. (only for H and W)

@jiafatom jiafatom merged commit 1e62db2 into onnx:master May 4, 2020
@jiafatom jiafatom deleted the conv_spacetond branch May 4, 2020 23:15
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants