The following error occurred when I tested my trained model. about pytorch_realtime_multi-person_pose_estimation HOT 4 OPEN

now I meet same error，

from pytorch_realtime_multi-person_pose_estimation.

In the demo code, update the lines to load the model as below:

model = get_model('vgg19')
wts_dict = torch.load(args.weight)
wts_dict_corrected = {k.replace('module.',''): v for k, v in wts_dict.items()}
model.load_state_dict(wts_dict_corrected)

from pytorch_realtime_multi-person_pose_estimation.

Hi,Sorry I dont know how to fix your problem.
But you trained success! I am also troubled with the train process.Could you give some ideas?
I train the model on a animal-pose datasets which has 20 keypoints and 15 keypoint-connection.
When I train, error log said: the shape of heatmap and vec not match to outputs of network(which is heatmap 19 and vec(paf) 38) ,I try to change the output channels of network, but got another error:RuntimeError: Given groups=1, weight of size [128, 185, 7, 7], expected input[8, 179, 46, 46] to have 185 channels, but got 179 channels instead.

from pytorch_realtime_multi-person_pose_estimation.

I think you downloaded a wrong weight.

from pytorch_realtime_multi-person_pose_estimation.