Article ID: 000088806 Content Type: Compatibility Last Reviewed: 05/20/2022

Unable to Convert FastSeg-Large Model into Intermediate Representation (IR) by Running “converter.py” Script

BUILT IN - ARTICLE INTRO SECOND COMPONENT
Summary

Install Torch* and Torchvision* libraries that are compatible with Model Converter.

Description
  • Downloaded FastSeg-large, Public Pre-Trained model from Open Model Zoo.
  • Ran converter.py to convert FastSeg-large model to Intermediate Representation (IR):
    python3 converter.py --name fastseg-large
  • Received error: cannot import name 'container_abcs' from 'torch._six'
Resolution
  1. Uninstall Torch and Torchvision:
    pip uninstall torch torchvision
  2. Install compatible version of Torch and Torchvision:
    pip install torch==1.8.0 torchvision==0.9.0
  3. Convert FastSeg-large model:
    python3 converter.py --name fastseg-large

Related Products

This article applies to 2 products