Got unexpected keyword argument "groups"

[zendevil]

python3 run.py
initialize
device: cuda
Loading weights: model-f46da743.pt
Downloading: "https://github.com/facebookresearch/WSL-Images/archive/master.zip" to /home/prikshet/.cache/torch/hub/master.zip
Traceback (most recent call last):
File "run.py", line 105, in
run(INPUT_PATH, OUTPUT_PATH, MODEL_PATH)
File "run.py", line 29, in run
model = MidasNet(model_path, non_negative=True)
File "/home/prikshet/midas/midas/midas_net.py", line 30, in init
self.pretrained, self.scratch = _make_encoder(features, use_pretrained)
File "/home/prikshet/midas/midas/blocks.py", line 6, in _make_encoder
pretrained = _make_pretrained_resnext101_wsl(use_pretrained)
File "/home/prikshet/midas/midas/blocks.py", line 26, in _make_pretrained_resnext101_wsl
resnet = torch.hub.load("facebookresearch/WSL-Images", "resnext101_32x8d_wsl")
File "/usr/local/lib/python3.6/dist-packages/torch/hub.py", line 354, in load
model = entry(*args, **kwargs)
File "/home/prikshet/.cache/torch/hub/facebookresearch_WSL-Images_master/hubconf.py", line 39, in resnext101_32x8d_wsl
return _resnext('resnext101_32x8d', Bottleneck, [3, 4, 23, 3], True, progress, **kwargs)
File "/home/prikshet/.cache/torch/hub/facebookresearch_WSL-Images_master/hubconf.py", line 23, in _resnext
model = ResNet(block, layers, **kwargs)
TypeError: init() got an unexpected keyword argument 'groups'

[ranftlr]

You might be using an incompatible version of PyTorch and/or Torchvision. What versions are you using? You can check with:

import torch
import torchvision

print(torch.__version__)
print(torchvision.__version__)

I tested it just now and I don't see the error with PyTorch 1.6.0 and Torchvision 0.7.0.

[ranftlr]

Closing due to inactivity.

[3dsf]

just a confirmation, it is related to the torchvision version
conda had installed something below 0.3.0 for me

---

TypeError: init() got an unexpected keyword argument 'groups'

torchvision version was too low (need 0.3.0 or higher) as per
facebookresearch/WSL-Images#3

---

if using a conda virtual environment, I'd recommend upgrading via pip
pip3 install torchvision --upgrade

[AlexanderStohr]

running on Windows

initial state:
torch=1.7.1+cpu
torchvision=0.2.2

this line failed for me - no newer version found:
pip3 install torchvision --upgrade

then i tried this one:
pip install torch==1.5.1+cpu torchvision==0.6.1+cpu -f https://download.pytorch.org/whl/torch_stable.html
(found it there: https://stackoverflow.com/questions/62513412/cannot-install-pip-install-torchvision )

due to some access restriction i finally added "--user" on above

final result:
torch=1.5.1+cpu
torchvision=0.6.1+cpu

and previous dangling operation is now past.

alternate combinations of CPU/CUDA, OS, installer and alikes can be found via the tool here:
https://pytorch.org/get-started/locally/

for more version pairings and indication what might match with your python version:
https://pypi.org/project/torchvision/