forked from mrq/DL-Art-School
44b89330c2
This is a checkpoint of a set of long tests with reduced-complexity networks. Some takeaways:
1) A full GAN using the resnet discriminator does appear to converge, but the quality is capped.
2) Likewise, a combination GAN/feature loss does not converge. The feature loss is optimized but
the model appears unable to fight the discriminator, so the G-loss steadily increases.
Going forwards, I want to try some bigger models. In particular, I want to change the generator
to increase complexity and capacity. I also want to add skip connections between the
disc and generator.
27 lines
514 B
YAML
27 lines
514 B
YAML
name: RRDB_ESRGAN_x4
|
|
suffix: ~ # add suffix to saved images
|
|
model: sr
|
|
distortion: sr
|
|
scale: 4
|
|
crop_border: ~ # crop border when evaluation. If None(~), crop the scale pixels
|
|
#gpu_ids: [0]
|
|
|
|
datasets:
|
|
test_1: # the 1st test dataset
|
|
name: set5
|
|
mode: LQ
|
|
batch_size: 1
|
|
dataroot_LQ: E:\4k6k\datasets\adrianna\full_extract
|
|
|
|
#### network structures
|
|
network_G:
|
|
which_model_G: RRDBNet
|
|
in_nc: 3
|
|
out_nc: 3
|
|
nf: 48
|
|
nb: 23
|
|
|
|
#### path
|
|
path:
|
|
pretrain_model_G: ../experiments/rrdb_blacked_gan_g.pth
|