tensorflow / tensorflow/models

MobileDet EdgeTPU model lacks post-processing step

Open
#8,769 8 comments 0 reactions 3 assignees View on GitHub

@pkulzc is already working on this.

Since Jul 2, 2020.

models:research:odapi type:feature
Dominant language
Python
Stars
77.7k
Forks
44.8k
PR merge metrics
No merged PRs in 30d

Description

Prerequisites

N/A
Please answer the following question for yourself before submitting an issue.

  • I checked to make sure that this feature has not been requested already.

1. The entire URL of the file you are using

https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md#pixel4-edge-tpu-models

2. Describe the feature you request

The CPU version of MobileDet has post-processing, but the EdgeTPU version does not. Moreover, I have been unable to convert the frozen graph to tflite to add the post-processing step via the provided export_ssd_tflite.py script with --add_postprocessing_op option. In short, I would like to request a MobileDet model with post-processing op for EdgeTPU, please.

3. Additional context

N/A

4. Are you willing to contribute it? (Yes or No)

Yes. I'm happy to convert it myself, but despite investigating the source code, I've been unable to reproduce the post-processing step and unable to convert the model to TFLite.

Contributor guide

Open the contributing guide

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.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.