Commit Graph

228 Commits

Author SHA1 Message Date
James Betker
eb12b5f887 Misc 2020-09-26 21:27:17 -06:00
James Betker
254cb1e915 More dataset integration work 2020-09-25 22:19:38 -06:00
James Betker
f211575e9d Save models before validation
Validation often fails with OOM, wasting hours of training time.
Save models first.
2020-09-16 08:17:17 -06:00
James Betker
c833bd1eac Misc changes 2020-09-15 20:57:59 -06:00
James Betker
747ded2bf7 Fixes to the spsr3
Some lessons learned:
- Biases are fairly important as a relief valve. They dont need to be everywhere, but
  most computationally heavy branches should have a bias.
- GroupNorm in SPSR is not a great idea. Since image gradients are represented
   in this model, normal means and standard deviations are not applicable. (imggrad
   has a high representation of 0).
- Don't fuck with the mainline of any generative model. As much as possible, all
   additions should be done through residual connections. Never pollute the mainline
   with reference data, do that in branches. It basically leaves the mode untrainable.
2020-09-09 15:28:14 -06:00
James Betker
c04f244802 More mods 2020-09-08 20:36:27 -06:00
James Betker
e6207d4c50 SPSR3 work
SPSR3 is meant to fix whatever is causing the switching units
inside of the newer SPSR architectures to fail and basically
not use the multiplexers.
2020-09-08 15:14:23 -06:00
James Betker
a18ece62ee Add updated spsr net for test 2020-09-07 17:01:48 -06:00
James Betker
e8613041c0 Add novograd optimizer 2020-09-06 17:27:08 -06:00
James Betker
6657a406ac Mods needed to support training a corruptor again:
- Allow original SPSRNet to have a specifiable block increment
- Cleanup
- Bug fixes in code that hasnt been touched in awhile.
2020-09-04 15:33:39 -06:00
James Betker
886d59d5df Misc fixes & adjustments 2020-09-01 07:58:11 -06:00
James Betker
0a9b85f239 Fix vgg_gn input_img_factor 2020-08-31 09:50:30 -06:00
James Betker
4b4d08bdec Enable testing in ExtensibleTrainer, fix it in SRGAN_model
Also compute fea loss for this.
2020-08-31 09:41:48 -06:00
James Betker
623f3b99b2 Stupid pathing.. 2020-08-26 17:58:24 -06:00
James Betker
80aa83bfd2 Try copytree for tb_logger again. 2020-08-26 17:55:02 -06:00
James Betker
b593d8e7c3 Save tb_logger to alt_path 2020-08-26 17:45:07 -06:00
James Betker
f35b3ad28f Fix val behavior for ExtensibleTrainer 2020-08-26 08:44:22 -06:00
James Betker
19487d9bbd Fix distributed launch for large distributed runs 2020-08-25 15:42:59 -06:00
James Betker
a65b07607c Reference network 2020-08-25 11:56:59 -06:00
James Betker
f9276007a8 More fixes to corrupt_fea 2020-08-23 17:52:18 -06:00
James Betker
dffc15184d More ExtensibleTrainer work
It runs now, just need to debug it to reach performance parity with SRGAN. Sweet.
2020-08-23 17:22:45 -06:00
James Betker
afdd93fbe9 Grey feature 2020-08-22 13:41:38 -06:00
James Betker
40bb0597bb misc 2020-08-18 08:50:24 -06:00
James Betker
0c98c61f4a Enable start_step to be specified 2020-08-15 18:34:59 -06:00
James Betker
2d205f52ac Unite spsr_arch switched gens
Found a pretty good basis model.
2020-08-12 17:04:45 -06:00
James Betker
bdaa67deb7 Misc 2020-08-12 08:46:15 -06:00
James Betker
1d5f4f6102 Crossgan 2020-08-07 21:03:39 -06:00
James Betker
3ab39f0d22 Several new spsr nets 2020-08-05 10:01:24 -06:00
James Betker
328afde9c0 Integrate SPSR into SRGAN_model
SPSR_model really isn't that different from SRGAN_model. Rather than continuing to re-implement
everything I've done in SRGAN_model, port the new stuff from SPSR over.

This really demonstrates the need to refactor SRGAN_model a bit to make it cleaner. It is quite the
beast these days..
2020-08-02 12:55:08 -06:00
James Betker
c8da78966b Substantial SPSR mods & fixes
- Added in gradient accumulation via mega-batch-factor
- Added AMP
- Added missing train hooks
- Added debug image outputs
- Cleaned up including removing GradientPenaltyLoss, custom SpectralNorm
- Removed all the custom discriminators
2020-08-02 10:45:24 -06:00
James Betker
f894ba8f98 Add SPSR_module
This is a port from the SPSR repo, it's going to need a lot of work to be properly integrated
but as of this commit it at least runs.
2020-08-01 22:02:54 -06:00
James Betker
eb11a08d1c Enable disjoint feature networks
This is done by pre-training a feature net that predicts the features
of HR images from LR images. Then use the original feature network
and this new one in tandem to work only on LR/Gen images.
2020-07-31 16:29:47 -06:00
James Betker
e37726f302 Add feature_model for training custom feature nets 2020-07-31 11:20:39 -06:00
James Betker
a7541b6d8d Fix illegal tb_logger use in distributed training 2020-07-23 09:14:01 -06:00
James Betker
dbf6147504 Add switched discriminator
The logic is that the discriminator may be incapable of providing a truly
targeted loss for all image regions since it has to be too generic
(basically the same argument for the switched generator). So add some
switches in! See how it works!
2020-07-22 20:52:59 -06:00
James Betker
8a9f215653 Huge set of mods to support progressive generator growth 2020-07-18 14:18:48 -06:00
James Betker
47a525241f Make attention norm optional 2020-07-18 07:24:02 -06:00
James Betker
8d061a2687 Add u-net discriminator with feature output 2020-07-16 10:10:09 -06:00
James Betker
1b1431133b Add DualOutputSRG
Also removes the old multi-return mechanism that Generators support.
Also fixes AttentionNorm.
2020-07-14 09:28:24 -06:00
James Betker
812c684f7d Update pixgan swap algorithm
- Swap multiple blocks in the image instead of just one. The discriminator was clearly
  learning that most blocks have one region that needs to be fixed.
- Relax block size constraints. This was in place to gaurantee that the discriminator
  signal was clean. Instead, just downsample the "loss image" with bilinear interpolation.
  The result is noisier, but this is actually probably healthy for the discriminator.
2020-07-10 15:56:14 -06:00
James Betker
5f2c722a10 SRG2 revival
Big update to SRG2 architecture to pull in a lot of things that have been learned:
- Use group norm instead of batch norm
- Initialize the weights on the transformations low like is done in RRDB rather than using the scalar. Models live or die by their early stages, and this ones early stage is pretty weak
- Transform multiplexer to use u-net like architecture.
- Just use one set of configuration variables instead of a list - flat networks performed fine in this regard.
2020-07-09 17:34:51 -06:00
James Betker
b2507be13c Fix up pixgan loss and pixdisc 2020-07-08 21:27:48 -06:00
James Betker
26a4a66d1c Bug fixes and new gan mechanism
- Removed a bunch of unnecessary image loggers. These were just consuming space and never being viewed
- Got rid of support of artificial var_ref support. The new pixdisc is what i wanted to implement then - it's much better.
- Add pixgan GAN mechanism. This is purpose-built for the pixdisc. It is intended to promote a healthy discriminator
- Megabatchfactor was applied twice on metrics, fixed that

Adds pix_gan (untested) which swaps a portion of the fake and real image with each other, then expects the discriminator
to properly discriminate the swapped regions.
2020-07-08 17:40:26 -06:00
James Betker
8a4eb8241d SRG3 work
Operates on top of a pre-trained SpineNET backbone (trained on CoCo 2017 with RetinaNet)

This variant is extremely shallow.
2020-07-07 13:46:40 -06:00
James Betker
a47a5dca43 Fix pixdisc bug 2020-07-05 21:57:52 -06:00
James Betker
188de5e15a Misc changes 2020-07-04 13:22:50 -06:00
James Betker
77d3765364 Fix new feature loss calc 2020-07-03 22:20:13 -06:00
James Betker
da4335c25e Add a feature-based validation test 2020-07-03 15:18:57 -06:00
James Betker
703dec4472 Add SpineNet & integrate with SRG
New version of SRG uses SpineNet for a switch backbone.
2020-07-03 12:07:31 -06:00
James Betker
ea9c6765ca Move train imports into init_dist 2020-07-02 15:11:21 -06:00
James Betker
c0bb123504 Misc changes 2020-07-01 11:28:23 -06:00
James Betker
2e3b6bad77 Log tensorboard directly into experiments directory 2020-06-18 11:33:02 -06:00
James Betker
45a900fafe Misc 2020-06-18 11:28:55 -06:00
James Betker
379b96eb55 Output histograms with SwitchedResidualGenerator
This also fixes the initialization weight for the configurable generator.
2020-06-16 15:54:37 -06:00
James Betker
6c27ddc9b5 Misc 2020-06-14 11:03:02 -06:00
James Betker
296135ec18 Add doResizeLoss to dataset
doResizeLoss has a 50% chance to resize the LQ image to 50% size,
then back to original size. This is useful to training a generator to
recover these lost pixel values while also being able to do
repairs on higher resolution images during training.
2020-06-08 11:27:06 -06:00
James Betker
ef5d8a0ed1 Misc 2020-06-05 21:01:50 -06:00
James Betker
726d1913ac Allow validating in batches, remove val size limit 2020-06-02 08:41:22 -06:00
James Betker
f1a1fd14b1 Introduce (untested) colab mode 2020-06-01 15:09:52 -06:00
James Betker
f6815df58b Misc 2020-05-27 08:04:47 -06:00
James Betker
3c2e5a0250 Apply fixes to resgen 2020-05-24 07:43:23 -06:00
James Betker
987cdad0b6 Misc mods 2020-05-23 21:09:38 -06:00
James Betker
9cde58be80 Make RRDB usable in the current iteration 2020-05-16 18:36:30 -06:00
James Betker
635c53475f Restore swapout models just before a checkpoint 2020-05-16 07:45:19 -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
bd4d478572 config changes 2020-05-15 07:41:18 -06:00
James Betker
8a514b9645 Misc changes 2020-05-14 20:45:38 -06:00
James Betker
037a5a3cdb Config updates 2020-05-13 09:20:28 -06:00
James Betker
602f86bfa4 Random config changes 2020-05-06 17:25:48 -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
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
9e1acfe396 Fixup upconv for the next attempt! 2020-05-01 19:56:14 -06:00
James Betker
03258445bc tblogger.. 2020-04-30 12:35:51 -06:00
James Betker
f027e888ed Clear out tensorboard on job restart. 2020-04-30 11:44:53 -06:00
James Betker
35bd1ecae4 Config changes for discriminator advantage run
Still going from high->low, discriminator discerns on low. Next up disc works on high.
2020-04-25 11:24:28 -06:00
James Betker
d95808f4ef Implement downsample GAN
This bad boy is for a workflow where you train a model on disjoint image sets to
downsample a "good" set of images like a "bad" set of images looks. You then
use that downsampler to generate a training set of paired images for supersampling.
2020-04-24 00:00:46 -06:00
James Betker
f4b33b0531 Some random fixes/adjustments 2020-04-22 00:38:53 -06:00
XintaoWang
037933ba66 mmsr 2019-08-23 21:42:47 +08:00