[QNN] Conv2D type checking for kernel per-channel scales. #4732
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
shoubhik
reviewed
Jan 17, 2020
src/relay/qnn/op/convolution.cc
Outdated
| CHECK(IsScalarType(types[5], DataType::Float(32))); // kernel_scale | ||
| // Kernel scale can be a vector of length output_channels or a scalar. | ||
| size_t axis = param->kernel_layout.find("O"); | ||
| AssignType(types[4], DataType::Float(32), weight->shape[axis], reporter); // kernel scale |
There was a problem hiding this comment.
should this be types[5]? there is a types[4] on line 59.
shoubhik
reviewed
Jan 17, 2020
src/relay/qnn/op/convolution.cc
Outdated
| @@ -57,7 +57,9 @@ bool QnnConv2DRel(const Array<Type>& types, int num_inputs, const Attrs& attrs, | |||
| CHECK(IsScalarType(types[2], DataType::Int(32))); // input_zero_point | |||
| CHECK(IsScalarType(types[3], DataType::Int(32))); // kernel_zero_point | |||
| CHECK(IsScalarType(types[4], DataType::Float(32))); // input_scale | |||
| CHECK(IsScalarType(types[5], DataType::Float(32))); // kernel_scale | |||
| // Kernel scale can be a vector of length output_channels or a scalar. | |||
| size_t axis = param->kernel_layout.find("O"); | |||
There was a problem hiding this comment.
Since it is a single char, better use 'O' instead of "O", from C++ perspective.
There was a problem hiding this comment.
Also we should check what happens if there is no "O", just to be safe.
|
@shoubhik Good catch about the index. Comments addressed. |
|
Thank you for ping, and sorry for missed this. I had a busy week. :) |
alexwong
pushed a commit
to alexwong/tvm
that referenced
this pull request
Feb 26, 2020
* [QNN] Conv2D type checking for kernel per-channel scales. * Address commments. * Address comments. * - Adding safety checks for downcasts. Co-authored-by: shoubhik <[email protected]>
alexwong
pushed a commit
to alexwong/tvm
that referenced
this pull request
Feb 28, 2020
* [QNN] Conv2D type checking for kernel per-channel scales. * Address commments. * Address comments. * - Adding safety checks for downcasts. Co-authored-by: shoubhik <[email protected]>
zhiics
pushed a commit
to neo-ai/tvm
that referenced
this pull request
Mar 2, 2020
* [QNN] Conv2D type checking for kernel per-channel scales. * Address commments. * Address comments. * - Adding safety checks for downcasts. Co-authored-by: shoubhik <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
As Title.
@FrozenGene @jackwish @shoubhik @yzhliu