How to produce the pretrained model for unguided network

[TruongKhang]

Hi, I have a few questions about your implementation.

1. I saw that you already had a pre-trained model for unguided_network.py. How to re-produce this pretrained model?
2. I didn't see you use the SoftPlus function in unguided_network.py. Can you explain it to me? In your paper, your unguided network was proposed to use this function.

[abdo-eldesokey]

Hi,
1- You can retrain "workspace/exp_unguided_depth" experiment.
2- Since it was pretrained and frozen during training the guided network, then the weights does not change and the SoftPlus is not needed.

[TruongKhang]

Hi, but the name of weights in the exp_ungided_network.py and the name of weights in the network_exp_unguided_depth.py are different. I'm curious about how can you load the pre-trained model of network_exp_unguided_depth.py into the model of unguided_network.py?