James Betker
62d3b6496b
Latent work checkpoint
2020-11-05 13:31:34 -07:00
James Betker
fd6cdba88f
RRDB with latent
2020-11-05 10:04:17 -07:00
James Betker
df47d6cbbb
More work in support of training flow networks in tandem with generators
2020-11-04 18:07:48 -07:00
James Betker
658a267bab
More work on SSIM/PSNR approximators
...
- Add a network that accomodates this style of approximator while retaining structure
- Migrate to SSIM approximation
- Add a tool to visualize how these approximators are working
- Fix some issues that came up while doign this work
2020-11-03 08:09:58 -07:00
James Betker
a51daacde2
Fix reporting of d_fake_diff for generators
2020-11-02 08:45:46 -07:00
James Betker
dcfe994fee
Add standalone srg2_classic
...
Trying to investigate how I was so misguided. I *thought* srg2 was considerably
better than RRDB in performance but am not actually seeing that.
2020-10-31 20:55:34 -06:00
James Betker
eb7df63592
Merge remote-tracking branch 'origin/gan_lab' into gan_lab
2020-10-31 11:09:32 -06:00
James Betker
c2866ad8d2
Disable debugging of comparable pingpong generations
2020-10-31 11:09:10 -06:00
James Betker
7303d8c932
Add psnr approximator
2020-10-31 11:08:55 -06:00
James Betker
565517814e
Restore SRG2
...
Going to try to figure out where SRG lost competitiveness to RRDB..
2020-10-30 14:01:56 -06:00
James Betker
74738489b9
Fixes and additional support for progressive zoom
2020-10-30 09:59:54 -06:00
James Betker
a3918fa808
Tecogan & other fixes
2020-10-30 00:19:58 -06:00
James Betker
b316078a15
Fix tecogan_losses fp16
2020-10-29 23:02:20 -06:00
James Betker
3791f95ad0
Enable RRDB to take in reference inputs
2020-10-29 11:07:40 -06:00
James Betker
7d38381d46
Add scaling to rrdb
2020-10-29 09:48:10 -06:00
James Betker
607ff3c67c
RRDB with bypass
2020-10-29 09:39:45 -06:00
James Betker
1655b9e242
Fix fast_forward teco loss bug
2020-10-28 17:49:54 -06:00
James Betker
515905e904
Add a min_loss that is DDP compatible
2020-10-28 15:46:59 -06:00
James Betker
f133243ac8
Extra logging for teco_resgen
2020-10-28 15:21:22 -06:00
James Betker
2ab5054d4c
Add noise to teco disc
2020-10-27 22:48:23 -06:00
James Betker
4dc16d5889
Upgrade tecogan_losses for speed
2020-10-27 22:40:15 -06:00
James Betker
ac3da0c5a6
Make tecogen functional
2020-10-27 21:08:59 -06:00
James Betker
10da206db6
Merge remote-tracking branch 'origin/gan_lab' into gan_lab
2020-10-27 20:59:59 -06:00
James Betker
9848f4c6cb
Add teco_resgen
2020-10-27 20:59:55 -06:00
James Betker
543c384a91
Merge remote-tracking branch 'origin/gan_lab' into gan_lab
2020-10-27 20:59:16 -06:00
James Betker
da53090ce6
More adjustments to support distributed training with teco & on multi_modal_train
2020-10-27 20:58:03 -06:00
James Betker
00bb568956
further checkpointify spsr_arch
2020-10-27 17:54:28 -06:00
James Betker
c2727a0150
Merge remote-tracking branch 'origin/gan_lab' into gan_lab
2020-10-27 15:24:19 -06:00
James Betker
2a3eec8fd7
Fix some distributed training snafus
2020-10-27 15:24:05 -06:00
James Betker
d923a62ed3
Allow SPSR to checkpoint
2020-10-27 15:23:20 -06:00
James Betker
11a9e223a6
Retrofit SPSR_arch so it is capable of accepting a ref
2020-10-27 11:14:36 -06:00
James Betker
8202ee72b9
Re-add original SPSR_arch
2020-10-27 11:00:38 -06:00
James Betker
231137ab0a
Revert RRDB back to original model
2020-10-27 10:25:31 -06:00
James Betker
1ce863849a
Remove temporary base_model change
2020-10-26 11:13:01 -06:00
James Betker
54accfa693
Merge remote-tracking branch 'origin/gan_lab' into gan_lab
2020-10-26 11:12:37 -06:00
James Betker
ff58c6484a
Fixes to unified chunk datasets to support stereoscopic training
2020-10-26 11:12:22 -06:00
James Betker
f857eb00a8
Allow tecogan losses to compute at 32px
2020-10-26 11:09:55 -06:00
James Betker
629b968901
ChainedGen 4x alteration
...
Increases conv window for teco_recurrent in the 4x case so all data
can be used.
base_model changes should be temporary.
2020-10-26 10:54:51 -06:00
James Betker
85c07f85d9
Update flownet submodule
2020-10-24 11:59:00 -06:00
James Betker
9c3d059ef0
Updates to be able to train flownet2 in ExtensibleTrainer
...
Only supports basic losses for now, though.
2020-10-24 11:56:39 -06:00
James Betker
1dbcbfbac8
Restore ChainedEmbeddingGenWithStructure
...
Still using this guy, after all
2020-10-24 11:54:52 -06:00
James Betker
7a75d10784
Arch cleanup
2020-10-23 09:35:33 -06:00
James Betker
646d6a621a
Support 4x zoom on ChainedEmbeddingGen
2020-10-23 09:25:58 -06:00
James Betker
e9c0b9f0fd
More adjustments to support multi-modal training
...
Specifically - looks like at least MSE loss cannot handle autocasted tensors
2020-10-22 16:49:34 -06:00
James Betker
76789a456f
Class-ify train.py and workon multi-modal trainer
2020-10-22 16:15:31 -06:00
James Betker
15e00e9014
Finish integration with autocast
...
Note: autocast is broken when also using checkpoint(). Overcome this by modifying
torch's checkpoint() function in place to also use autocast.
2020-10-22 14:39:19 -06:00
James Betker
d7ee14f721
Move to torch.cuda.amp (not working)
...
Running into OOM errors, needs diagnosing. Checkpointing here.
2020-10-22 13:58:05 -06:00
James Betker
3e3d2af1f3
Add multi-modal trainer
2020-10-22 13:27:32 -06:00
James Betker
40dc2938e8
Fix multifaceted chain gen
2020-10-22 13:27:06 -06:00
James Betker
43c4f92123
Collapse progressive zoom candidates into the batch dimension
...
This contributes a significant speedup to training this type of network
since losses can operate on the entire prediction spectrum at once.
2020-10-21 22:37:23 -06:00
James Betker
680d635420
Enable ExtensibleTrainer to skip steps when state keys are missing
2020-10-21 22:22:28 -06:00
James Betker
d1175f0de1
Add FFT injector
2020-10-21 22:22:00 -06:00
James Betker
1ef559d7ca
Add a ChainedEmbeddingGen which can be simueltaneously used with multiple training paradigms
2020-10-21 22:21:51 -06:00
James Betker
931aa65dd0
Allow recurrent losses to be weighted
2020-10-21 16:59:44 -06:00
James Betker
5753e77d67
ChainedGen: Output debugging information on blocks
2020-10-21 16:36:23 -06:00
James Betker
3c6e600e48
Add capacity for models to self-report visuals
2020-10-21 11:08:03 -06:00
James Betker
dca5cddb3b
Add bypass to ChainedEmbeddingGen
2020-10-21 11:07:45 -06:00
James Betker
a63bf2ea2f
Merge remote-tracking branch 'origin/gan_lab' into gan_lab
2020-10-19 15:26:11 -06:00
James Betker
76e4f0c086
Restore test.py for use as standalone validator
2020-10-19 15:26:07 -06:00
James Betker
1b1ca297f8
Fix recurrent=None bug in ChainedEmbeddingGen
2020-10-19 15:25:12 -06:00
James Betker
b28e4d9cc7
Add spread loss
...
Experimental loss that peaks around 0.
2020-10-19 11:31:19 -06:00
James Betker
981d64413b
Support validation over a custom injector
...
Also re-enable PSNR
2020-10-19 11:01:56 -06:00
James Betker
668cafa798
Push correct patch of recurrent embedding to upstream image, rather than whole thing
2020-10-18 22:39:52 -06:00
James Betker
7df378a944
Remove separated vgg discriminator
...
Checkpointing happens inline instead. Was a dumb idea..
Also fixes some loss reporting issues.
2020-10-18 12:10:24 -06:00
James Betker
c709d38cd5
Fix memory leak with recurrent loss
2020-10-18 10:22:10 -06:00
James Betker
552e70a032
Get rid of excessive checkpointed disc params
2020-10-18 10:09:37 -06:00
James Betker
6a0d5f4813
Add a checkpointable discriminator
2020-10-18 09:57:47 -06:00
James Betker
9ead2c0a08
Multiscale training in!
2020-10-17 22:54:12 -06:00
James Betker
e706911c83
Fix spinenet bug
2020-10-17 20:20:36 -06:00
James Betker
b008a27d39
Spinenet should allow bypassing the initial conv
...
This makes feeding in references for recurrence easier.
2020-10-17 20:16:47 -06:00
James Betker
c1c9c5681f
Swap recurrence
2020-10-17 08:40:28 -06:00
James Betker
6141aa1110
More recurrence fixes for chainedgen
2020-10-17 08:35:46 -06:00
James Betker
cf8118a85b
Allow recurrence to specified for chainedgen
2020-10-17 08:32:29 -06:00
James Betker
fc4c064867
Add recurrent support to chainedgenwithstructure
2020-10-17 08:31:34 -06:00
James Betker
d4a3e11ab2
Don't use several stages of spinenet_arch
...
These are used for lower outputs which I am not using
2020-10-17 08:28:37 -06:00
James Betker
d1c63ae339
Go back to torch's DDP
...
Apex was having some weird crashing issues.
2020-10-16 20:47:35 -06:00
James Betker
d856378b2e
Add ChainedGenWithStructure
2020-10-16 20:44:36 -06:00
James Betker
617d97e19d
Add ChainedEmbeddingGen
2020-10-15 23:18:08 -06:00
James Betker
c4543ce124
Set post_transform_block to None where applicable
2020-10-15 17:20:42 -06:00
James Betker
6f8705e8cb
SSGSimpler network
2020-10-15 17:18:44 -06:00
James Betker
eda75c9779
Cleanup fixes
2020-10-15 10:13:17 -06:00
James Betker
920865defb
Arch work
2020-10-15 10:13:06 -06:00
James Betker
1f20d59c31
Revert big switch back
2020-10-14 11:03:34 -06:00
James Betker
24792bdb4f
Codebase cleanup
...
Removed a lot of legacy stuff I have no intent on using again.
Plan is to shape this repo into something more extensible (get it? hah!)
2020-10-13 20:56:39 -06:00
James Betker
e620fc05ba
Mods to support video processing with teco networks
2020-10-13 20:47:05 -06:00
James Betker
17d78195ee
Mods to SRG to support returning switch logits
2020-10-13 20:46:37 -06:00
James Betker
cc915303a5
Fix SPSR calls into SwitchComputer
2020-10-13 10:14:47 -06:00
James Betker
bdf4c38899
Merge remote-tracking branch 'origin/gan_lab' into gan_lab
...
# Conflicts:
# codes/models/archs/SwitchedResidualGenerator_arch.py
2020-10-13 10:12:26 -06:00
James Betker
9a5d6162e9
Add the "BigSwitch"
2020-10-13 10:11:10 -06:00
James Betker
8014f050ac
Clear metrics properly
...
Holy cow, what a PITA bug.
2020-10-13 10:07:49 -06:00
James Betker
4d52374e60
Merge remote-tracking branch 'origin/gan_lab' into gan_lab
2020-10-12 17:43:51 -06:00
James Betker
731700ab2c
checkpoint in ssg
2020-10-12 17:43:28 -06:00
James Betker
ca523215c6
Fix recurrent std in arch
2020-10-12 17:42:32 -06:00
James Betker
05377973bf
Allow initial recurrent input to be specified (optionally)
2020-10-12 17:36:43 -06:00
James Betker
597b6e92d6
Add ssgr1 recurrence
2020-10-12 17:18:19 -06:00
James Betker
d7d7590f3e
Fix constant injector - wasn't working in test
2020-10-12 10:36:30 -06:00
James Betker
ce163ad4a9
Update SSGdeep
2020-10-12 10:22:08 -06:00
James Betker
3409d88a1c
Add PANet arch
2020-10-12 10:20:55 -06:00
James Betker
a9c2e97391
Constant injector and teco fixes
2020-10-11 08:20:07 -06:00
James Betker
e785029936
Mods needed to support SPSR archs with teco gan
2020-10-10 22:39:55 -06:00
James Betker
120072d464
Add constant injector
2020-10-10 21:50:23 -06:00
James Betker
f99812e14d
Fix tecogan_losses errors
2020-10-10 20:30:14 -06:00
James Betker
3a5b23b9f7
Alter teco_losses to feed a recurrent input in as separate
2020-10-10 20:21:09 -06:00
James Betker
0d30d18a3d
Add MarginRemoval injector
2020-10-09 20:35:56 -06:00
James Betker
0011d445c8
Fix loss indexing
2020-10-09 20:20:51 -06:00
James Betker
202eb11fdc
For element loss added
2020-10-09 19:51:44 -06:00
James Betker
fe50d6f9d0
Fix attention images
2020-10-09 19:21:55 -06:00
James Betker
7e777ea34c
Allow tecogan to be used in process_video
2020-10-09 19:21:43 -06:00
James Betker
58d8bf8f69
Add network architecture built for teco
2020-10-09 08:40:14 -06:00
James Betker
afe6af88af
Fix attention print issue
2020-10-08 18:34:00 -06:00
James Betker
4c85ee51a4
Converge SSG architectures into unified switching base class
...
Also adds attention norm histogram to logging
2020-10-08 17:23:21 -06:00
James Betker
1eb516d686
Fix more distributed bugs
2020-10-08 14:32:45 -06:00
James Betker
fba29d7dcc
Move to apex distributeddataparallel and add switch all_reduce
...
Torch's distributed_data_parallel is missing "delay_allreduce", which is
necessary to get gradient checkpointing to work with recurrent models.
2020-10-08 11:20:05 -06:00
James Betker
c174ac0fd5
Allow tecogan to support generators that only output a tensor (instead of a list)
2020-10-08 09:26:25 -06:00
James Betker
969bcd9021
Use local checkpoint in SSG
2020-10-08 08:54:46 -06:00
James Betker
c93dd623d7
Tecogan losses work
2020-10-07 23:11:58 -06:00
James Betker
c96f5b2686
Import switched_conv as a submodule
2020-10-07 23:10:54 -06:00
James Betker
c352c8bce4
More tecogan fixes
2020-10-07 12:41:17 -06:00
James Betker
1c44d395af
Tecogan work
...
Its training! There's still probably plenty of bugs though..
2020-10-07 09:03:30 -06:00
James Betker
e9d7371a61
Add concatenate injector
2020-10-07 09:02:42 -06:00
James Betker
8a7e993aea
Merge remote-tracking branch 'origin/gan_lab' into gan_lab
2020-10-06 20:41:58 -06:00
James Betker
1e415b249b
Add tag that can be applied to prevent parameter training
2020-10-06 20:39:49 -06:00
James Betker
2f2e3f33f8
StackedSwitchedGenerator_5lyr
2020-10-06 20:39:32 -06:00
James Betker
6217b48e3f
Fix spsr_arch bug
2020-10-06 20:38:47 -06:00
James Betker
cffc596141
Integrate flownet2 into codebase, add teco visual debugs
2020-10-06 20:35:39 -06:00
James Betker
e4b89a172f
Reduce spsr7 memory usage
2020-10-05 22:05:56 -06:00
James Betker
4111942ada
Support attention deferral in deep ssgr
2020-10-05 19:35:55 -06:00
James Betker
840927063a
Work on tecogan losses
2020-10-05 19:35:28 -06:00
James Betker
2875822024
SPSR9 arch
...
takes some of the stuff I learned with SGSR yesterday and applies it to spsr
2020-10-05 08:47:51 -06:00
James Betker
51044929af
Don't compute attention statistics on multiple generator invocations of the same data
2020-10-05 00:34:29 -06:00
James Betker
e760658fdb
Another fix..
2020-10-04 21:08:00 -06:00
James Betker
a890e3a9c0
Fix geometric loss not handling 0 index
2020-10-04 21:05:01 -06:00
James Betker
c3ef8a4a31
Stacked switches - return a tuple
2020-10-04 21:02:24 -06:00
James Betker
13f97e1e97
Add recursive loss
2020-10-04 20:48:15 -06:00
James Betker
ffd069fd97
Lots of SSG work
...
- Checkpointed pretty much the entire model - enabling recurrent inputs
- Added two new models for test - adding depth (again) and removing SPSR (in lieu of the new losses)
2020-10-04 20:48:08 -06:00
James Betker
aca2c7ab41
Full checkpoint-ize SSG1
2020-10-04 18:24:52 -06:00
James Betker
e3294939b0
Revert "SSG: offer option to use BN-based attention normalization"
...
Didn't work. Oh well.
This reverts commit 5cd2b37591
.
2020-10-03 17:54:53 -06:00
James Betker
5cd2b37591
SSG: offer option to use BN-based attention normalization
...
Not sure how this is going to work, lets try it.
2020-10-03 16:16:19 -06:00
James Betker
9b4ed82093
Get rid of unused convs in spsr7
2020-10-03 11:36:26 -06:00
James Betker
3561cc164d
Fix up fea_loss calculator (for validation)
...
Not sure how this was working in regular training mode, but it
was failing in DDP.
2020-10-03 11:19:20 -06:00
James Betker
6c9718ad64
Don't log if you aren't 0 rank
2020-10-03 11:14:13 -06:00
James Betker
922b1d76df
Don't record visuals when not on rank 0
2020-10-03 11:10:03 -06:00
James Betker
8197fd646f
Don't accumulate losses for metrics when the loss isn't a tensor
2020-10-03 11:03:55 -06:00
James Betker
19a4075e1e
Allow checkpointing to be disabled in the options file
...
Also makes options a global variable for usage in utils.
2020-10-03 11:03:28 -06:00
James Betker
dd9d7b27ac
Add more sophisticated mechanism for balancing GAN losses
2020-10-02 22:53:42 -06:00
James Betker
39865ca3df
TOTAL_loss, dumbo
2020-10-02 21:06:10 -06:00
James Betker
4e44fcd655
Loss accumulator fix
2020-10-02 20:55:33 -06:00
James Betker
567b4d50a4
ExtensibleTrainer - don't compute backward when there is no loss
2020-10-02 20:54:06 -06:00
James Betker
146a9125f2
Modify geometric & translational losses so they can be used with embeddings
2020-10-02 20:40:13 -06:00
James Betker
e30a1443cd
Change sw2 refs
2020-10-02 09:01:18 -06:00
James Betker
e38716925f
Fix spsr8 class init
2020-10-02 09:00:18 -06:00
James Betker
35469f08e2
Spsr 8
2020-10-02 08:58:15 -06:00
James Betker
aa4fd89018
resnext with groupnorm
2020-10-01 15:49:28 -06:00
James Betker
8beaa47933
resnext discriminator
2020-10-01 11:48:14 -06:00
James Betker
55f2764fef
Allow fixup50 to be used as a discriminator
2020-10-01 11:28:18 -06:00
James Betker
7986185fcb
Change 'mod_step' to 'every'
2020-10-01 11:28:06 -06:00
James Betker
d9ae970fd9
SSG update
2020-10-01 11:27:51 -06:00
James Betker
e3053e4e55
Exchange SpsrNet for SpsrNetSimplified
2020-09-30 17:01:04 -06:00
James Betker
66d4512029
Fix up translational equivariance loss so it's ready for prime time
2020-09-30 12:01:00 -06:00
James Betker
896b4f5be2
Revert "spsr7 adjustments"
...
This reverts commit 9fee1cec71
.
2020-09-29 18:30:41 -06:00
James Betker
9fee1cec71
spsr7 adjustments
2020-09-29 17:19:59 -06:00
James Betker
dc8f3b24de
Don't let duplicate keys be used for injectors and losses
2020-09-29 16:59:44 -06:00
James Betker
0b5a033503
spsr7 + cleanup
...
SPSR7 adds ref onto spsr6, makes more "common sense" mods.
2020-09-29 16:59:26 -06:00
James Betker
f9b83176f1
Fix bugs in extensibletrainer
2020-09-28 22:09:42 -06:00
James Betker
db52bec4ab
spsr6
...
This is meant to be a variant of SPSR5 that harkens
back to the simpler earlier architectures that do not
have embeddings or ref_ inputs, but do have deep
multiplexers. It does, however, use some of the new
conjoin mechanisms.
2020-09-28 22:09:27 -06:00
James Betker
7e240f2fed
Recurrent / teco work
2020-09-28 22:06:56 -06:00
James Betker
aeaf185314
Add RCAN
2020-09-27 16:00:41 -06:00
James Betker
4d29b7729e
Model arch cleanup
2020-09-27 11:18:45 -06:00
James Betker
31641d7f63
Add ImagePatchInjector and TranslationalLoss
2020-09-26 21:25:32 -06:00
James Betker
d8621e611a
BackboneSpineNoHead takes ref
2020-09-26 21:25:04 -06:00
James Betker
5a27187c59
More mods to accomodate new dataset
2020-09-25 22:45:57 -06:00
James Betker
6d0490a0e6
Tecogan implementation work
2020-09-25 16:38:23 -06:00
James Betker
ce4613ecb9
Finish up single_image_dataset work
...
Sweet!
2020-09-25 16:37:54 -06:00
James Betker
ea565b7eaf
More fixes
2020-09-24 17:51:52 -06:00
James Betker
553917a8d1
Fix torchvision import bug
2020-09-24 17:38:34 -06:00
James Betker
58886109d4
Update how spsr arches do attention to conform with sgsr
2020-09-24 16:53:54 -06:00
James Betker
9a50a7966d
SiLU doesnt support inplace
2020-09-23 21:09:13 -06:00
James Betker
eda0eadba2
Use custom SiLU
...
Torch didnt have this before 1.7
2020-09-23 21:05:06 -06:00
James Betker
05963157c1
Several things
...
- Fixes to 'after' and 'before' defs for steps (turns out they werent working)
- Feature nets take in a list of layers to extract. Not fully implemented yet.
- Fixes bugs with RAGAN
- Allows real input into generator gan to not be detached by param
2020-09-23 11:56:36 -06:00
James Betker
4ab989e015
try again..
2020-09-22 18:27:52 -06:00
James Betker
3b6c957194
Fix? it again?
2020-09-22 18:25:59 -06:00
James Betker
7b60d9e0d8
Fix? cosine loss
2020-09-22 18:18:35 -06:00
James Betker
2e18c4c22d
Add CosineEmbeddingLoss to F
2020-09-22 17:10:29 -06:00
James Betker
f40beb5460
Add 'before' and 'after' defs to injections, steps and optimizers
2020-09-22 17:03:22 -06:00
James Betker
419f77ec19
Some new backbones
2020-09-21 12:36:49 -06:00
James Betker
9429544a60
Spinenet: implementation without 4x downsampling right off the bat
2020-09-21 12:36:30 -06:00
James Betker
53a5657850
Fix SSGR
2020-09-20 19:07:15 -06:00
James Betker
17c569ea62
Add geometric loss
2020-09-20 16:24:23 -06:00
James Betker
17dd99b29b
Fix bug with discriminator noise addition
...
It wasn't using the scale and was applying the noise to the
underlying state variable.
2020-09-20 12:00:27 -06:00
James Betker
3138f98fbc
Allow discriminator noise to be injected at the loss level, cleans up configs
2020-09-19 21:47:52 -06:00
James Betker
e9a39bfa14
Recursively detach all outputs, even if they are nested in data structures
2020-09-19 21:47:34 -06:00
James Betker
fe82785ba5
Add some new architectures to ssg
2020-09-19 21:47:10 -06:00
James Betker
b83f097082
Get rid of get_debug_values from RRDB, rectify outputs
2020-09-19 21:46:36 -06:00
James Betker
e0bd68efda
Add ImageFlowInjector
2020-09-19 10:07:00 -06:00
James Betker
e2a146abc7
Add in experiments hook
2020-09-19 10:05:25 -06:00
James Betker
9a17ade550
Some convenience adjustments to ExtensibleTrainer
2020-09-17 21:05:32 -06:00
James Betker
9963b37200
Add a new script for loading a discriminator network and using it to filter images
2020-09-17 13:30:32 -06:00
James Betker
723754c133
Update attention debugger outputting for SSG
2020-09-16 13:09:46 -06:00
James Betker
0918430572
SSG network
...
This branches off of SPSR. It is identical but substantially reduced
in complexity. It's intended to be my long term working arch.
2020-09-15 20:59:24 -06:00
James Betker
6deab85b9b
Add BackboneEncoderNoRef
2020-09-15 16:55:38 -06:00
James Betker
d0321ca5de
Don't load amp state dict if amp is disabled
2020-09-14 15:21:42 -06:00
James Betker
ccf8438001
SPSR5
...
This is SPSR4, but the multiplexers have access to the output of the transformations
for making their decision.
2020-09-13 20:10:24 -06:00
James Betker
5b85f891af
Only log the name of the first network in the total_loss training set
2020-09-12 16:07:09 -06:00
James Betker
fb595e72a4
Supporting infrastructure in ExtensibleTrainer to train spsr4
...
Need to be able to train 2 nets in one step: the backbone will be entirely separate
with its own optimizer (for an extremely low LR).
This functionality was already present, just not implemented correctly.
2020-09-11 22:57:06 -06:00
James Betker
4e44bca611
SPSR4
...
aka - return of the backbone! I'm tired of massively overparameterized generators
with pile-of-shit multiplexers. Let's give this another try..
2020-09-11 22:55:37 -06:00
James Betker
19896abaea
Clean up old SwitchedSpsr arch
...
It didn't work anyways, so why not?
2020-09-11 16:09:28 -06:00
James Betker
50ca17bb0a
Feature mode -> back to LR fea
2020-09-11 13:09:55 -06:00
James Betker
1086f0476b
Fix ref branch using fixed filters
2020-09-11 08:58:35 -06:00
James Betker
8c469b8286
Enable memory checkpointing
2020-09-11 08:44:29 -06:00
James Betker
5189b11dac
Add combined dataset for training across multiple datasets
2020-09-11 08:44:06 -06:00
James Betker
313424d7b5
Add new referencing discriminator
...
Also extend the way losses work so that you can pass
parameters into the discriminator from the config file
2020-09-10 21:35:29 -06:00
James Betker
9e5aa166de
Report the standard deviation of ref branches
...
This patch also ups the contribution
2020-09-10 16:34:41 -06:00
James Betker
668bfbff6d
Back to best arch for spsr3
2020-09-10 14:58:14 -06:00
James Betker
992b0a8d98
spsr3 with conjoin stage as part of the switch
2020-09-10 09:11:37 -06:00
James Betker
e0fc5eb50c
Temporary commit - noise
2020-09-09 17:12:52 -06:00
James Betker
00da69d450
Temporary commit - ref
2020-09-09 17:09:44 -06:00
James Betker
df59d6c99d
More spsr3 mods
...
- Most branches get their own noise vector now.
- First attention branch has the intended sole purpose of raw image processing
- Remove norms from joiner block
2020-09-09 16:46:38 -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
0ffac391c1
SPSR with ref joining
2020-09-09 11:17:07 -06:00
James Betker
3027e6e27d
Enable amp to be disabled
2020-09-09 10:45:59 -06:00
James Betker
c04f244802
More mods
2020-09-08 20:36:27 -06:00
James Betker
dffbfd2ec4
Allow SRG checkpointing to be toggled
2020-09-08 15:14:43 -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
5606e8b0ee
Fix SRGAN_model/fullimgdataset compatibility 1
2020-09-08 11:34:35 -06:00
James Betker
22c98f1567
Move MultiConvBlock to arch_util
2020-09-08 08:17:27 -06:00
James Betker
146ace0859
CSNLN changes (removed because it doesnt train well)
2020-09-08 08:04:16 -06:00
James Betker
f43df7f5f7
Make ExtensibleTrainer compatible with process_video
2020-09-08 08:03:41 -06:00
James Betker
a18ece62ee
Add updated spsr net for test
2020-09-07 17:01:48 -06:00
James Betker
55475d2ac1
Clean up unused archs
2020-09-07 11:38:11 -06:00
James Betker
e8613041c0
Add novograd optimizer
2020-09-06 17:27:08 -06:00
James Betker
b1238d29cb
Fix trainable not applying to discriminators
2020-09-05 20:31:26 -06:00
James Betker
21ae135f23
Allow Novograd to be used as an optimizer
2020-09-05 16:50:13 -06:00
James Betker
912a4d9fea
Fix srg computer bug
2020-09-05 07:59:54 -06:00
James Betker
0dfd8eaf3b
Support injectors that run in eval only
2020-09-05 07:59:45 -06:00
James Betker
44c75f7642
Undo SRG change
2020-09-04 17:32:16 -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
bfdfaab911
Checkpoint RRDB
...
Greatly reduces memory consumption with a low performance penalty
2020-09-04 15:32:00 -06:00
James Betker
696242064c
Use tensor checkpointing to drastically reduce memory usage
...
This comes at the expense of computation, but since we can use much larger
batches, it results in a net speedup.
2020-09-03 11:33:36 -06:00
James Betker
365813bde3
Add InterpolateInjector
2020-09-03 11:32:47 -06:00
James Betker
d90c96e55e
Fix greyscale injector
2020-09-02 10:29:40 -06:00
James Betker
8b52d46847
Interpreted feature loss to extensibletrainer
2020-09-02 10:08:24 -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
b2091cb698
feamod fix
2020-08-30 08:08:49 -06:00
James Betker
a56e906f9c
train HR feature trainer
2020-08-29 09:27:48 -06:00
James Betker
0e859a8082
4x spsr ref (not workin)
2020-08-29 09:27:18 -06:00
James Betker
25832930db
Update loss with lr crossgan
2020-08-26 17:57:22 -06:00
James Betker
cbd5e7a986
Support old school crossgan in extensibletrainer
2020-08-26 17:52:35 -06:00
James Betker
8a6a2e6e2e
Rev3 of the full image ref arch
2020-08-26 17:11:01 -06:00
James Betker
f35b3ad28f
Fix val behavior for ExtensibleTrainer
2020-08-26 08:44:22 -06:00
James Betker
434ed70a9a
Wrap vgg disc
2020-08-25 18:14:45 -06:00
James Betker
83f2f8d239
more debugging
2020-08-25 18:12:12 -06:00
James Betker
3f60281da7
Print when wrapping
2020-08-25 18:08:46 -06:00
James Betker
bae18c05e6
wrap disc grad
2020-08-25 17:58:20 -06:00
James Betker
f85f1e21db
Turns out, can't do that
2020-08-25 17:18:52 -06:00
James Betker
935a735327
More dohs
2020-08-25 17:05:16 -06:00
James Betker
53e67bdb9c
Distribute get_grad_no_padding
2020-08-25 17:03:18 -06:00
James Betker
2f706b7d93
I an inept.
2020-08-25 16:42:59 -06:00
James Betker
8bae0de769
ffffffffffffffffff
2020-08-25 16:41:01 -06:00
James Betker
1fe16f71dd
Fix bug reporting spsr gan weight
2020-08-25 16:37:45 -06:00
James Betker
96586d6592
Fix distributed d_grad
2020-08-25 16:06:27 -06:00
James Betker
09a9079e17
Check rank before doing image logging.
2020-08-25 16:00:49 -06:00
James Betker
a1800f45ef
Fix for referencingmultiplexer
2020-08-25 15:43:12 -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
0005c56cd4
dbg
2020-08-23 17:43:03 -06:00
James Betker
4bb5b3c981
corfea debugging
2020-08-23 17:39:02 -06:00
James Betker
7713cb8df5
Corrupted features in srgan
2020-08-23 17:32:03 -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
e59e712e39
More ExtensibleTrainer work
2020-08-22 13:08:33 -06:00
James Betker
f40545f235
ExtensibleTrainer work
2020-08-22 08:24:34 -06:00
James Betker
a498d7b1b3
Report l_g_gan_grad before weight multiplication
2020-08-20 11:57:53 -06:00
James Betker
9d77a4db2e
Allow initial temperature to be specified to SPSR net for inference
2020-08-20 11:57:34 -06:00
James Betker
24bdcc1181
Let SwitchedSpsr transform count be specified
2020-08-18 09:10:25 -06:00
James Betker
74cdaa2226
Some work on extensible trainer
2020-08-18 08:49:32 -06:00
James Betker
868d0aa442
Undo early dim reduction on grad branch for SPSR_arch
2020-08-14 16:23:42 -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
3d0ece804b
SPSR LR2
2020-08-12 08:45:49 -06:00
James Betker
ab04ca1778
Extensible trainer (in progress)
2020-08-12 08:45:23 -06:00
James Betker
cb316fabc7
Use LR data for image gradient prediction when HR data is disjoint
2020-08-10 15:00:28 -06:00
James Betker
f0e2816239
Denoise attention maps
2020-08-10 14:59:58 -06:00
James Betker
59aba1daa7
LR switched SPSR arch
...
This variant doesn't do conv processing at HR, which should save
a ton of memory in inference. Lets see how it works.
2020-08-10 13:03:36 -06:00
James Betker
4e972144ae
More attention fixes for switched_spsr
2020-08-07 21:11:50 -06:00
James Betker
d02509ef97
spsr_switched missing import
2020-08-07 21:05:29 -06:00
James Betker
887806ffa0
Finish up spsr_switched
2020-08-07 21:03:48 -06:00
James Betker
1d5f4f6102
Crossgan
2020-08-07 21:03:39 -06:00
James Betker
fd7b6ca0a9
Comptue gan_grad_branch....
2020-08-06 12:11:40 -06:00
James Betker
30b16d5235
Update how branch GAN grad is disseminated
2020-08-06 11:13:02 -06:00
James Betker
1f21c02f8b
Add cross-compare discriminator
2020-08-06 08:56:21 -06:00
James Betker
be272248af
More RAGAN fixes
2020-08-05 16:47:21 -06:00
James Betker
26a6a5d512
Compute grad GAN loss against both the branch and final target, simplify pixel loss
...
Also fixes a memory leak issue where we weren't detaching our loss stats when
logging them. This stabilizes memory usage substantially.
2020-08-05 12:08:15 -06:00
James Betker
299ee13988
More RAGAN fixes
2020-08-05 11:03:06 -06:00
James Betker
b8a4df0a0a
Enable RAGAN in SPSR, retrofit old RAGAN for efficiency
2020-08-05 10:34:34 -06:00
James Betker
3ab39f0d22
Several new spsr nets
2020-08-05 10:01:24 -06:00
James Betker
3c0a2d6efe
Fix grad branch debug out
2020-08-04 16:43:43 -06:00
James Betker
ec2a795d53
Fix multistep optimizer (feeding from wrong config params)
2020-08-04 16:42:58 -06:00
James Betker
4bfbdaf94f
Don't recompute generator outputs for D in standard operation
...
Should significantly improve training performance with negligible
results differences.
2020-08-04 11:28:52 -06:00
James Betker
11b227edfc
Whoops
2020-08-04 10:30:40 -06:00
James Betker
6d25bcd5df
Apply fixes to grad discriminator
2020-08-04 10:25:13 -06:00
James Betker
c7e5d3888a
Add pix_grad_branch loss to metrics
2020-08-03 16:21:05 -06:00
James Betker
0d070b47a7
Add simplified SPSR architecture
...
Basically just cleaning up the code, removing some bad conventions,
and reducing complexity somewhat so that I can play around with
this arch a bit more easily.
2020-08-03 10:25:37 -06:00
James Betker
47e24039b5
Fix bug that makes feature loss run even when it is off
2020-08-02 20:37:51 -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
f33ed578a2
Update how attention_maps are created
2020-08-01 20:23:46 -06:00
James Betker
c139f5cd17
More torch 1.6 fixes
2020-07-31 17:03:20 -06:00
James Betker
a66fbb32b6
Fix fixed_disc DataParallel issue
2020-07-31 16:59:23 -06:00
James Betker
8dd44182e6
Fix scale torch warning
2020-07-31 16:56:04 -06:00
James Betker
bcebed19b7
Fix pixdisc bugs
2020-07-31 16:38:14 -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
6e086d0c20
Fix fixed_disc
2020-07-31 15:07:10 -06:00
James Betker
d5fa059594
Add capability to have old discriminators serve as feature networks
2020-07-31 14:59:54 -06:00
James Betker
6b45b35447
Allow multi_step_lr_scheduler to load a new LR schedule when restoring state
2020-07-31 11:21:11 -06:00
James Betker
e37726f302
Add feature_model for training custom feature nets
2020-07-31 11:20:39 -06:00
James Betker
7629cb0e61
Add FDPL Loss
...
New loss type that can replace PSNR loss. Works against the frequency domain
and focuses on frequency features loss during hr->lr conversion.
2020-07-30 20:47:57 -06:00
James Betker
85ee64b8d9
Turn down feadisc intensity
...
Honestly - this feature is probably going to be removed soon, so backwards
compatibility is not a huge deal anymore.
2020-07-27 15:28:55 -06:00
James Betker
ebb199e884
Get rid of safety valve (probably being encountered in val)
2020-07-26 22:51:59 -06:00
James Betker
d09ed4e5f7
Misc fixes
2020-07-26 22:44:24 -06:00
James Betker
c54784ae9e
Fix feature disc log item error
2020-07-26 22:25:59 -06:00
James Betker
9a8f227501
Allow separate dataset to pushed in for GAN-only training
2020-07-26 21:44:45 -06:00
James Betker
b06e1784e1
Fix SRG4 & switch disc
...
"fix". hehe.
2020-07-25 17:16:54 -06:00
James Betker
e6e91a1d75
Add SRG4
...
Back to the idea that maybe what we need is a hybrid
approach between pure switches and RDB.
2020-07-24 20:32:49 -06:00
James Betker
3320ad685f
Fix mega_batch_factor not set for test
2020-07-24 12:26:44 -06:00
James Betker
c50cce2a62
Add an abstract, configurabler weight scheduling class and apply it to the feature weight
2020-07-23 17:03:54 -06:00
James Betker
9ccf771629
Fix feature validation, wrong device
...
Only shows up in distributed training for some reason.
2020-07-23 10:16:34 -06:00
James Betker
bba283776c
Enable find_unused_parameters for DistributedDataParallel
...
attention_norm has some parameters which are not used to compute grad,
which is causing failures in the distributed case.
2020-07-23 09:08:13 -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
106b8da315
Assert that temperature is set properly in eval mode.
2020-07-22 20:50:59 -06:00
James Betker
c74b9ee2e4
Add a way to disable grad on portions of the generator graph to save memory
2020-07-22 11:40:42 -06:00
James Betker
e3adafbeac
Add convert_model.py and a hacky way to add extra layers to a model
2020-07-22 11:39:45 -06:00
James Betker
7f7e17e291
Update feature discriminator further
...
Move the feature/disc losses closer and add a feature computation layer.
2020-07-20 20:54:45 -06:00
James Betker
46aa776fbb
Allow feature discriminator unet to only output closest layer to feature output
2020-07-19 19:05:08 -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
ad97a6a18a
Progressive SRG first check-in
2020-07-18 07:23:26 -06:00
James Betker
b08b1cad45
Fix feature decay
2020-07-16 23:27:06 -06:00
James Betker
3e7a83896b
Fix pixgan debugging issues
2020-07-16 11:45:19 -06:00
James Betker
a1bff64d1a
More fixes
2020-07-16 10:48:48 -06:00
James Betker
240f254263
More loss fixes
2020-07-16 10:45:50 -06:00
James Betker
6cfa67d831
Fix featuredisc broadcast error
2020-07-16 10:18:30 -06:00
James Betker
8d061a2687
Add u-net discriminator with feature output
2020-07-16 10:10:09 -06:00
James Betker
0c4c388e15
Remove dualoutputsrg
...
Good idea, didn't pan out.
2020-07-16 10:09:24 -06:00
James Betker
4bcc409fc7
Fix loadSRG2 typo
2020-07-14 10:20:53 -06:00
James Betker
1e4083a35b
Apply temperature mods to all SRG models
...
(Honestly this needs to be base classed at this point)
2020-07-14 10:19:35 -06:00
James Betker
7659bd6818
Fix temperature equation
2020-07-14 10:17:14 -06:00
James Betker
853468ef82
Allow legacy state_dicts in srg2
2020-07-14 10:03:45 -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