onnx / onnx/models

ShuffleNet Model wrong shape input dimension 0 for Reshape operator

Open
#229 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
9.8k
Forks
1.6k
PR merge metrics
No merged PRs in 30d

Description

id: OC2_DUMMY_1
kind: Initializer
type: int64[5]
[
    1,
    4,
    28,
    56,
    56
]

When running the model w/ multiple images at once, batch dimension of Data input will of course change. ONNX Reshape allows for that by allowing value of -1 for a dimension (means: "infer the shape"). In this particular case shape[0] is to be -1, not 1. Same for the other Reshape after Transpose.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Locate the ShuffleNet model referenced in the issue and inspect its Reshape initializers, including OC2_DUMMY_1 and the later Reshape after Transpose. Verify how the model handles multiple input images, then update the batch dimension from 1 to -1 in both shapes. Confirm the model accepts a batch size other than one.

Written by the indexing model from the issue text.

Assessment

Domain
machine-learning
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
38/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.