Alex J. Champandard
|
87304c93a6
|
Use traditional learning rate decaying rather than fast-restarts, works better when training continuously adapting GAN.
|
9 years ago |
|
Alex J. Champandard
|
3809e9b02a
|
Add loading of images into a buffer, using multiple fragments per JPG loaded. Works well with larger datasets like OpenImages, fully GPU bound.
|
9 years ago |
|
Alex J. Champandard
|
619fad7f3c
|
Add loading parameters from saved models. Clean up learning-rate code.
|
9 years ago |
|
Alex J. Champandard
|
f978f058dd
|
Add descriptions for command-line arguments.
|
9 years ago |
|
Alex J. Champandard
|
d8c9292b80
|
Tweak code for generalized super-resolution.
|
9 years ago |
|
Alex J. Champandard
|
9558a11397
|
Add command-line for super-resolving images without training.
|
9 years ago |
|
Alex J. Champandard
|
03a6813e95
|
Enable adversarial loss during training.
|
9 years ago |
|
Alex J. Champandard
|
6542f435b4
|
Stabilize the training, tuned the architecture.
|
9 years ago |
|
Alex J. Champandard
|
b52530cf20
|
Add support for PRELU (default), optional load/save.
|
9 years ago |
|
Alex J. Champandard
|
4d82c7f157
|
Improve console output.
|
9 years ago |
|
Alex J. Champandard
|
8770113f9a
|
Add support for loading/saving models. Fixed regressions.
|
9 years ago |
|
Alex J. Champandard
|
58e1bed4a6
|
Improve display and code structure.
|
9 years ago |
|
Alex J. Champandard
|
749b467f94
|
Switch to sub-pixel deconvolution layer.
|
9 years ago |
|
Alex J. Champandard
|
9f13695050
|
Add discriminator network too.
|
9 years ago |
|
Alex J. Champandard
|
3f24714039
|
Add residual blocks to generator.
|
9 years ago |
|
Alex J. Champandard
|
058e3d3b9e
|
Add simple but working prototype, perceptual loss.
|
9 years ago |
