Error(s) in loading state_dict for PSMNet:

[evgdobr]

Hi,
i had faced with next issue while running step 2.4. Train the stereo model:

python3 ./psmnet/finetune_3d.py --maxdisp 192 \
>      --model stackhourglass \
>      --datapath ./KITTI/object/training/ \
>      --split_file ./KITTI/object/train.txt \
>      --epochs 300 \
>      --lr_scale 50 \
>      --loadmodel ./pretrained_sceneflow.tar \
>      --savemodel ./psmnet/kitti_3d/  --btrain 12
./psmnet/kitti_3d/training.log
[2019-09-20 15:17:27 finetune_3d.py:77] INFO     load model ./pretrained_sceneflow.tar
Traceback (most recent call last):
  File "./psmnet/finetune_3d.py", line 79, in <module>
    model.load_state_dict(state_dict['state_dict'])
  File "/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py", line 845, in load_state_dict
    self.__class__.__name__, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for PSMNet:
        Missing key(s) in state_dict: "feature_extraction.firstconv.0.0.weight", "feature_extraction.firstconv.0.1.weight", "feature_extraction.firstconv.0.1.bias", "feature_extraction.firstconv.0.1.running_mean", "feature_extraction.firstconv.0.1.running_var", "feature_extraction.firstconv.2.0.weight", "feature_extraction

[mileyan]

Hi Evgdobr,

You need to use python 2.7 when you run PSMNET code.

[evgdobr]

thanks,
already done it and it works.