"QuantizeWeights" returns a corrupted model when the used model has a large amount of weight constants

Package: Inference Engine

Open

2.6.X

How to reproduce:
1. Open the attached "IN-150605_bugReport.zip" project
2. Open the "SampleScene"
3. Enter Play mode
4. Observe the Console

Actual result: A "KeyNotFoundException: The given key '152' was not present in the dictionary." error is thrown in the Console, indicating that "QuantizeWeights" previously returned a corrupt model
Expected result: No errors are thrown in the Console

Reproducible with: 2.4.0 (6000.0.0f1), 2.6.1 (6000.7.0a4)
Not reproducible with: 2.3.0 (6000.0.0f1)

Reproducible on: Windows 11
Not reproducible on: No other environments tested

Prioritise and Follow this issue

Have a workaround or additional info about this issue?

Issues you vote on will appear here

"QuantizeWeights" returns a corrupted model when the used model has a large amount of weight constants

Package: Inference Engine

Statuses

Open

Resolution Notes

Description

Prioritise and Follow this issue

Have a workaround or additional info about this issue?

"QuantizeWeights" returns a corrupted model when the used model has a large amount of weight constants

Package: Inference Engine

Statuses

Open

Resolution Notes

Description

Prioritise and Follow this issue

Have a workaround or additional info about this issue?