I have tried to test Mask-Rcnn Resnet50 benchmarks on INTEL platforms using DL Workbench both on the DevCloud and local machine. Neither of them could successfully convert to INT8 model.
The Failed Code:137 was shown when using DevCloud optimized DL Workbench as below:
On the other hand, the local machine simply hang in the calibration stage.
Is there any way I could resolve this issue?
Thanks in advance,
Thank you for reaching out to us.
Please provide us with the following information to help us investigate this matter:
- Your model
- Annotated dataset (you may refer to Dataset Types)
- DL workbench image version
Thank you for your question. If you need any additional information from Intel, please submit a new question as this thread is no longer being monitored.