Commit Graph

96 Commits

Author SHA1 Message Date
James Betker
89c71293ce IDEA update 2020-05-19 09:35:26 -06:00
James Betker
9cde58be80 Make RRDB usable in the current iteration 2020-05-16 18:36:30 -06:00
James Betker
b95c4087d1 Allow an alt_path for saving models and states 2020-05-16 09:10:51 -06:00
James Betker
f911ef0d3e Add corruptor_usage_probability
Governs how often a corruptor is used, vs feeding uncorrupted images.
2020-05-16 09:05:43 -06:00
James Betker
635c53475f Restore swapout models just before a checkpoint 2020-05-16 07:45:19 -06:00
James Betker
877be4d88c README update 2020-05-15 14:03:44 -06:00
James Betker
a33ec3e22b Fix skips & images samples
- Makes skip connections between the generator and discriminator more
  extensible by adding additional configuration options for them and supporting
  1 and 0 skips.
- Places the temp/ directory with sample images from the training process appear
  in the training directory instead of the codes/ directory.
2020-05-15 13:50:49 -06:00
James Betker
cdf641e3e2 Remove working options from repo 2020-05-15 07:50:55 -06:00
James Betker
bd4d478572 config changes 2020-05-15 07:41:18 -06:00
James Betker
79593803f2 biggan arch, initial work (not implemented) 2020-05-15 07:40:45 -06:00
James Betker
61ed51d9e4 Improve corruptor logic: switch corruptors randomly 2020-05-14 23:14:32 -06:00
James Betker
d72e154442 Allow more LQ than GT images in corrupt mode 2020-05-14 20:46:20 -06:00
James Betker
8a514b9645 Misc changes 2020-05-14 20:45:38 -06:00
James Betker
a946483f1c Fix discriminator noise floor 2020-05-14 20:45:06 -06:00
James Betker
c8ab89d243 Add model swapout
Model swapout is a feature where, at specified intervals,
a random D and G model will be swapped in place for the
one being trained. After a short period of time, this model
is swapped back out. This is intended to increase training
diversity.
2020-05-13 16:53:38 -06:00
James Betker
c336d32fd3 Config updates 2020-05-13 15:27:49 -06:00
James Betker
5bcf187fb6 Disable LMDB support
It doesn't play nice with multiple dataroots and I don't
really see any need to continue support since I'm not
testing it.
2020-05-13 15:27:33 -06:00
James Betker
e36f22e14a Allow "corruptor" network to be specified
This network is just a fixed (pre-trained) generator
that performs a corruption transformation that the
generator-in-training is expected to undo alongside
SR.
2020-05-13 15:26:55 -06:00
James Betker
f389025b53 Change ResGen noise feature
It now injects noise directly into the input filters, rather than a
pure noise filter. The pure noise filter was producing really
poor results (and I'm honestly not quite sure why).
2020-05-13 09:22:06 -06:00
James Betker
343af70a8d Add code for compiling model to torchscript
I want to be able to export it to other formats too in the future.
2020-05-13 09:21:13 -06:00
James Betker
585b05e66b Cap test workers at 10 2020-05-13 09:20:45 -06:00
James Betker
037a5a3cdb Config updates 2020-05-13 09:20:28 -06:00
James Betker
fc3ec8e3a2 Add a noise floor to th discriminator noise factor 2020-05-13 09:19:22 -06:00
James Betker
5d1b4caabf Allow noise to be injected at the generator inputs for resgen 2020-05-12 16:26:29 -06:00
James Betker
06d18343f7 Allow noise to be added to discriminator inputs 2020-05-12 16:25:38 -06:00
James Betker
9210a62f58 Add rotating log buffer to trainer
Should stabilize stats output.
2020-05-12 10:09:45 -06:00
James Betker
f217216c81 Implement ResGenv2
Implements a ResGenv2 architecture which slightly increases the complexity
of the final output layer but causes it to be shared across all skip outputs.
2020-05-12 10:09:15 -06:00
James Betker
1596a98493 Get rid of skip layers from vgg disc 2020-05-12 10:08:12 -06:00
James Betker
c540244789 Config file update 2020-05-12 10:07:58 -06:00
James Betker
62a97c53d1 Handle tuple-returning generators in test 2020-05-11 11:15:26 -06:00
James Betker
f994466289 Initialize test dataloader with a worker count proportional to the batch size. 2020-05-10 10:49:37 -06:00
James Betker
ef48e819aa Allow resgen to have a conditional number of upsamples applied to it 2020-05-10 10:48:37 -06:00
James Betker
8969a3ce70 Add capability to start at arbitrary frames 2020-05-10 10:48:05 -06:00
James Betker
03351182be Use amp in SR_model for inference 2020-05-07 21:45:33 -06:00
James Betker
dbca0d328c Fix multi-lq bug 2020-05-06 23:16:35 -06:00
James Betker
aa0305def9 Resnet discriminator overhaul
It's been a tough day figuring out WTH is going on with my discriminators.
It appears the raw FixUp discriminator can get into an "defective" state where
they stop trying to learn and just predict as close to "0" D_fake and D_real as
possible. In this state they provide no feedback to the generator and never
recover. Adding batch norm back in seems to fix this so it must be some sort
of parameterization error.. Should look into fixing this in the future.
2020-05-06 17:27:30 -06:00
James Betker
602f86bfa4 Random config changes 2020-05-06 17:25:48 -06:00
James Betker
574e7e882b Fix up OOM issues when running a disjoint D update ratio and megabatches 2020-05-06 17:25:25 -06:00
James Betker
eee9d6d9ca Support skip connections in vgg arch discriminator. 2020-05-06 17:24:34 -06:00
James Betker
5c1832e124 Add support for multiple LQ paths
I want to be able to specify many different transformations onto
the target data; the model should handle them all. Do this by
allowing multiple LQ paths to be selected and the dataset class
selects one at random.
2020-05-06 17:24:17 -06:00
James Betker
3cd85f8073 Implement ResGen arch
This is a simpler resnet-based generator which performs mutations
on an input interspersed with interpolate-upsampling. It is a two
part generator:
1) A component that "fixes" LQ images with a long string of resnet
    blocks. This component is intended to remove compression artifacts
    and other noise from a LQ image.
2) A component that can double the image size. The idea is that this
    component be trained so that it can work at most reasonable
    resolutions, such that it can be repeatedly applied to itself to
    perform multiple upsamples.

The motivation here is to simplify what is being done inside of RRDB.
I don't believe the complexity inside of that network is justified.
2020-05-05 11:59:46 -06:00
James Betker
9f4581aacb Fix megabatch scaling, log low and med-res gen images 2020-05-05 08:34:57 -06:00
James Betker
3b4e54c4c5 Add support for passthrough disc/gen
Add RRDBNetXL, which performs processing at multiple image sizes.
Add DiscResnet_passthrough, which allows passthrough of image at different sizes for discrimination.
Adjust the rest of the repo to allow generators that return more than just a single image.
2020-05-04 14:01:43 -06:00
James Betker
44b89330c2 Support inference across batches, support inference on cpu, checkpoint
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.
2020-05-04 08:48:25 -06:00
James Betker
9c7debe75c Add colab option 2020-05-02 17:47:25 -06:00
James Betker
832f3587c5 Turn off EVDR (so we dont need the weird convs) 2020-05-02 17:47:14 -06:00
James Betker
8341bf7646 Enable megabatching 2020-05-02 17:46:59 -06:00
James Betker
61d3040cf5 Add doCrop into LQGT 2020-05-02 17:46:30 -06:00
James Betker
9e1acfe396 Fixup upconv for the next attempt! 2020-05-01 19:56:14 -06:00
James Betker
7eaabce48d Full resnet corrupt, no BN
And it works! Thanks fixup..
2020-04-30 19:17:30 -06:00