search

NVIDIA / flownet2-pytorch

Pytorch implementation of FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks
Other
3.14k stars 741 forks source link

NameError: name 'inference_loader' is not defined #250

Closed slokaanugu closed 3 years ago

slokaanugu commented 3 years ago

Hi,

I am getting NameError: name 'inference_loader' is not defined error.

I am running the below command:

python3 main.py --inference --model FlowNet2 --save_flow --inference_dataset MpiSintelClean --inference_dataset_root /home/user/sp/prediction/flownet2-pytorch/mpi-sintel/clean --resume checkpoint/FlowNet2_checkpoint.pth.tar

The folder structure of my mpi-sintel is as follows:

mpi-sintel -clean 1.bundler

  1. test 2a. final 2b. clean 3.flow_code

Where Am I going wrong? For the inference_dataset_root argument, I tried giving the below paths but all are giving the same error: 1./home/user/sp/prediction/flownet2-pytorch/mpi-sintel/clean/test

  1. /home/user/sp/prediction/flownet2-pytorch/mpi-sintel/clean/final
  2. /home/user/sp/prediction/flownet2-pytorch/mpi-sintel/clean/final/wall
izhimahu commented 3 years ago

I have the same problem. Have you solved it?