class documentation
class MMRealSR(nn.Module): (source)
Constructor: MMRealSR(num_in_ch, num_out_ch, scale, num_feat, ...)
Networks consisting of Residual in Residual Dense Block, which is used in ESRGAN. v2.1
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks.
We extend ESRGAN for scale x2 and scale x1. Note: This is one option for scale 1, scale 2 in RRDBNet. We first employ the pixel-unshuffle (an inverse operation of pixelshuffle to reduce the spatial size and enlarge the channel size before feeding inputs into the main ESRGAN architecture.
Parameters | |
num | Channel number of inputs. |
num | Channel number of outputs. |
num | Channel number of intermediate features. Default: 64 |
num | Block number in the trunk network. Defaults: 23 |
num | Channels for each growth. Default: 32. |
Method | __init__ |
Undocumented |
Method | forward |
Undocumented |
Class Variable | hyperparameters |
Undocumented |
Instance Variable | am |
Undocumented |
Instance Variable | body |
Undocumented |
Instance Variable | conv |
Undocumented |
Instance Variable | conv |
Undocumented |
Instance Variable | conv |
Undocumented |
Instance Variable | conv |
Undocumented |
Instance Variable | conv |
Undocumented |
Instance Variable | conv |
Undocumented |
Instance Variable | dd |
Undocumented |
Instance Variable | de |
Undocumented |
Instance Variable | lrelu |
Undocumented |
Instance Variable | num |
Undocumented |
Instance Variable | num |
Undocumented |
Instance Variable | scale |
Undocumented |