search

mapillary / seamseg

Seamless Scene Segmentation
BSD 3-Clause "New" or "Revised" License
289 stars 53 forks source link

Can we run inference on a non CUDA machine? #36

Open hoominchu opened 10 months ago

hoominchu commented 10 months ago

I am trying to run inference using the pre-trained models on a MacOS machine and getting an AssertionError thrown:

    assert self.local_world_size > 0
           ^^^^^^^^^^^^^^^^^^^^^^^^^
AssertionError

This is the command I am trying to run: python3 -m torch.distributed.launch --nproc_per_node=0 test_panoptic.py --meta ../models/SeamSeg-R50-Vistas/metadata.bin --log_dir ../logs/ ../models/SeamSeg-R50-Vistas/config.ini ../models/SeamSeg-R50-Vistas/seamseg_r50_vistas.tar ../input/ ../output/

Any help will be much appreciated!