skumra / robotic-grasping

Antipodal Robotic Grasping using GR-ConvNet. IROS 2020.
Other
455 stars 119 forks source link

Given groups=1, weight of size [32,1,9,9],expected input[1,4 ,300,300] to have 1 0 channels,but got 4 channels instead #74

Open chuh195 opened 6 months ago

chuh195 commented 6 months ago

Hello, when I train jacquard_samples data, I can run through the training team model by myself, but applying the model you have already trained gives me an error: RuntimerError in grcnn: Given groups=1, weight of size [32,1,9,9],expected input[1,4 ,300,300] to have 1 0 channels,but got 4 channels instead, I would like to ask how this problem arises and what should I try to solve it!