ResNet2d.py

[kiristern]

Create model.

• make sure the data that is provided to the network fits the input layer (you may or may not need some reshape functions applied to input data)
• the output of the network fits the shape of ground truth vector (again, you may or may not need some reshape functions applied to output here)

[kiristern]

references: train_gan_solutions
basset: convnet.lua

[kiristern]

for now just think about the input and output of network, that would be enough to get the training and evaluation loop running