wyfdodogod

> I think it is ok to just keep the output of calculate_pitch_yaw_roll and the input of tf.cos in the same unit, in spite of whether the unit is degree...

does that mean any model trained in torch version 3 and above need to be retrained in torch 2 and then safe to be converted to caffe

@autocyz ty for the explaination, as you mentioned using torch 0.2 and torchvision 0.1.8, i solved the problem more specifically torch0.2 post1 using post2 and 3 might raise error causing...