mrq/DL-Art-School
1
3
Fork 5
Code Issues Pull Requests Packages Projects Releases Wiki Activity
1,678 Commits 1 Branch 0 Tags 5.6 MiB
99de63a922
Commit Graph

1644 Commits

Author SHA1 Message Date
James Betker
f126040da2 Undo noise first 2021-08-05 23:24:38 -06:00
James Betker
908ef5495f Add noise first to audio_aug 2021-08-05 23:22:44 -06:00
James Betker
d6007c6de1 dataset fixes 2021-08-05 23:12:59 -06:00
James Betker
3ca51e80b2 Only fix weird path bug in windows 2021-08-05 22:21:25 -06:00
James Betker
70dcd1107f Fix byol_model_wrapper to function with audio inputs 2021-08-05 22:20:22 -06:00
James Betker
f86df53ce0 Export extract_byol_model as a function 2021-08-05 22:15:26 -06:00
James Betker
89d15c9e74 Move gpt-tts back to lucidrains implementation 
Much better performance.
 2021-08-05 22:15:13 -06:00
James Betker
d120e1aa99 Add audio augmentation to wavfile_dataset, utility to test audio similary 2021-08-05 22:14:49 -06:00
James Betker
c0f61a2e15 Rework how DVAE tokens are ordered 
It might make more sense to have top tokens, then bottom tokens
with top tokens having different discretized values.
 2021-08-05 07:07:17 -06:00
James Betker
4017236ba9 Fix up inference for gpt_tts 2021-08-05 06:46:30 -06:00
James Betker
5037220ac7 Mods to support contrastive learning on audio files 2021-08-05 05:57:04 -06:00
James Betker
341f28dd82 It works! 2021-08-04 20:07:51 -06:00
James Betker
36c7c1fbdb Fix training flow for NEXT TOKEN prediction instead of same token prediction 
doh
 2021-08-04 10:28:09 -06:00
James Betker
d9936df363 Add gpt_tts dataset and implement inference 
- Adds a script which preprocesses quantized mels given a DVAE
- Adds a dataset which can consume preprocessed qmels
- Reworks GPT TTS to consume the outputs of that dataset (removes logic to add padding and start/end tokens)
- Adds inference to gpt_tts
 2021-08-04 00:44:04 -06:00
James Betker
4c98b9703f Get dalle-style TTS to "work" 2021-08-03 21:08:27 -06:00
James Betker
2814307eee Alterations to support VQVAE on mel spectrograms 2021-08-01 07:54:21 -06:00
James Betker
965f6e6b52 Fixes to weight_decay in adamw 2021-07-31 15:58:41 -06:00
James Betker
0c9e75bc69 Improvements to GptTts 2021-07-31 15:57:57 -06:00
James Betker
31ee9ae262 Checkin 2021-07-30 23:07:35 -06:00
James Betker
dadc54795c Add gpt_tts 2021-07-27 20:33:30 -06:00
James Betker
398185e109 More work on wave-diffusion 2021-07-27 05:36:17 -06:00
James Betker
49e3b310ea Allow audio sample rate interpolation for faster training 2021-07-26 17:44:06 -06:00
James Betker
96e90e7047 Add support for a gaussian-diffusion-based wave tacotron 2021-07-26 16:27:31 -06:00
James Betker
97d7cbbc34 Additional work for audio xformer (which doesnt really do a great job) 2021-07-23 10:58:14 -06:00
James Betker
2325e7a88c Allow inference for vqvae 2021-07-20 10:40:05 -06:00
James Betker
d81386c1be Mods to support vqvae in audio mode (1d) 2021-07-20 08:36:46 -06:00
James Betker
5584cfcc7a tacotron2 work 2021-07-14 21:41:57 -06:00
James Betker
fe0c699ced Various fixes 2021-07-14 00:08:42 -06:00
James Betker
be2745f42d Add waveglow & inference capabilities to audio generator 2021-07-08 23:07:36 -06:00
James Betker
1ff434218e tacotron2, ready for prime time! 2021-07-08 22:13:44 -06:00
James Betker
86fd3ad7fd Initial checkin of nvidia tacotron model & dataset 
These two are tested, full support for training to come.
 2021-07-06 11:11:35 -06:00
James Betker
3801d5d55e diffusion surfin' 2021-07-06 09:36:52 -06:00
James Betker
afa41f1804 Allow hq color jittering and corruptions that are not included in the corruption factor 2021-06-30 09:44:46 -06:00
James Betker
6fd16ea9c8 Add meta-anomaly detection, colorjitter augmentation 2021-06-29 13:41:55 -06:00
James Betker
46e9f62be0 Add unet with latent guide 
This is a diffusion network that uses both a LQ image
and a reference sample HQ image that is compressed into
a latent vector to perform upsampling

The hope is that we can steer the upsampling network
with sample images.
 2021-06-26 11:02:58 -06:00
James Betker
0ded106562 Merge remote-tracking branch 'origin/master' 2021-06-25 13:16:28 -06:00
James Betker
a57ed8e960 Various mods to support better jpeg image filtering 2021-06-25 13:16:15 -06:00
James Betker
61e7ca39cd
Update image_folder_dataset.py 2021-06-25 11:48:31 -06:00
James Betker
a0ef07ddb8
Create unet_latent_guide.py 2021-06-25 11:25:14 -06:00
James Betker
e7890dc0ba Misc fixes for diffusion nets 2021-06-21 10:38:07 -06:00
James Betker
8e3a33e001 Fix a bug where non-rank-0 is computing FID before all images are saved. 2021-06-16 16:27:09 -06:00
James Betker
68cbbed886 Add some cool diffusion testing scripts 2021-06-16 16:26:36 -06:00
James Betker
ae8de0cb9d fid saving images across all rank fix 2021-06-15 10:31:07 -06:00
James Betker
6a75bd0777 Another fix 2021-06-14 09:51:44 -06:00
James Betker
54bff35171 Fix issue where eval was not being used by all ddp processes 2021-06-14 09:50:04 -06:00
James Betker
60079a1572 Fix saver in distributed mode 2021-06-14 09:41:06 -06:00
James Betker
545f2db170 Distributed FID dataset across processes 2021-06-14 09:33:44 -06:00
James Betker
6b32c87dcb Try to make diffusion fid more deterministic 2021-06-14 09:27:43 -06:00
James Betker
5b4f86293f Add FID evaluator for diffusion models 2021-06-14 09:14:30 -06:00
James Betker
9cfe840872 Attempt to fix syncing multiple times when doing gradient accumulation 2021-06-13 14:30:30 -06:00
James Betker
1cd75dfd33 Fix ddp bug 2021-06-13 10:25:23 -06:00
James Betker
3e3ad7825f Add support for training an EMA network alongside the main networks 2021-06-12 21:01:41 -06:00
James Betker
696f320820 Get rid of feature networks 2021-06-11 20:50:07 -06:00
James Betker
65c474eecf Various changes to fix testing 2021-06-11 15:31:10 -06:00
James Betker
220f11a5e4 Half channel sizes in cifar_resnet 2021-06-09 17:06:37 -06:00
James Betker
aea12e1b9c Fix cat eval hack 2021-06-09 17:05:11 -06:00
James Betker
9b5f4abb91 Add fade in for hard switch 2021-06-07 18:15:09 -06:00
James Betker
108c5d829c Fix dropout norm 2021-06-07 16:13:23 -06:00
James Betker
438217094c Also debug distribution of switch 2021-06-07 15:36:07 -06:00
James Betker
44b09e5f20 Amplify dropout rate 2021-06-07 15:20:53 -06:00
James Betker
f0d4eb9182 Fixor 2021-06-07 11:58:36 -06:00
James Betker
c456a60466 Another go at fixing nan 2021-06-07 11:51:43 -06:00
James Betker
1c574c5bd1 Attempt to fix nan 2021-06-07 11:43:42 -06:00
James Betker
eda796985b Try out dropout norm 2021-06-07 11:33:33 -06:00
James Betker
6c6e82406e Pass a corruption factor through the dataset into the upsampling network 
The intuition is this will help guide the network to make better informed decisions
about how it performs upsampling based on how it perceives the underlying content.

(I'm giving up on letting networks detect their own quality - I'm not convinced it is
actually feasible)
 2021-06-07 09:13:54 -06:00
James Betker
2ad2b56438 Don't do wandb except on rank 0 2021-06-06 16:52:07 -06:00
James Betker
7c5478bc2c Formatting issue with gdi 2021-06-06 16:35:37 -06:00
James Betker
061dbcd458 Another fix to anorm 2021-06-06 15:09:49 -06:00
James Betker
9a6991e461 Fix switch norm average 2021-06-06 15:04:28 -06:00
James Betker
57e1a6a0f2 cifar: add hard routing 
Also mods switched_routing to support non-pixular inputs
 2021-06-06 14:53:43 -06:00
James Betker
692e9c417b Support diffusion unet 2021-06-06 13:57:22 -06:00
James Betker
a0158ebc69 Simplify cifar resnet further for faster training 2021-06-06 10:02:24 -06:00
James Betker
75567a9814 Only head norm removed 2021-06-05 23:29:11 -06:00
James Betker
65d0376b90 Re-add normalization at the tail of the RRDB 2021-06-05 23:04:05 -06:00
James Betker
184e887122 Remove rrdb normalization 2021-06-05 21:39:19 -06:00
James Betker
f5e75602b9 Add regular attention to cifar_resnet 2021-06-05 21:34:07 -06:00
James Betker
16cd92acd5 hack 2021-06-05 14:23:41 -06:00
James Betker
af52751d6b Fix device error 2021-06-05 14:21:32 -06:00
James Betker
5f0cc65f3b Register branched resnet properly 2021-06-05 14:19:03 -06:00
James Betker
fb405d9ef1 CIFAR stuff 
- Extract coarse labels for the CIFAR dataset
- Add simple resnet that branches lower layers based on coarse labels
- Some other cleanup
 2021-06-05 14:16:02 -06:00
James Betker
80d4404367 A few fixes: 
- Output better prediction of xstart from eps
- Support LossAwareSampler
- Support AdamW
 2021-06-05 13:40:32 -06:00
James Betker
fa908a6a15 Fix wandb import issue 2021-06-04 23:27:15 -06:00
James Betker
103a88506e Log eval to wandb 2021-06-04 23:23:20 -06:00
James Betker
7d45132f60 fdsa 2021-06-04 21:26:54 -06:00
James Betker
6c8c8087d5 asdf 2021-06-04 21:24:48 -06:00
James Betker
e6c537824a Allow validation for ce 2021-06-04 21:21:04 -06:00
James Betker
7c251af7a8 Support cifar100 with resnet 2021-06-04 17:29:07 -06:00
James Betker
bf811f80c1 GD mods & fixes 
- Report variational loss separately
- Report model prediction from injector
- Log these things
- Use respacing like guided diffusion
 2021-06-04 17:13:16 -06:00
James Betker
6084915af8 Support gaussian diffusion models 
Adds support for GD models, courtesy of some maths from openai.

Also:
- Fixes requirement for eval{} even when it isn't being used
- Adds support for denormalizing an imagenet norm
 2021-06-02 21:47:32 -06:00
James Betker
45bc76ba92 Fixes and mods to support training classifiers on imagenet 2021-06-01 17:25:24 -06:00
James Betker
f129eaa39e Clean up byol a bit 
- Remove option to aug in dataset (there's really no reason for this now that kornia works on GPU on windows)
- Other stufff
 2021-05-24 21:35:46 -06:00
James Betker
6649ef2dae Add zipfilesdataset 2021-05-24 21:35:00 -06:00
James Betker
1a2b9fa130 Get rid of old byol net wrapping 
Simplifies and makes this usable with DLAS' multi-gpu trainer
 2021-04-27 12:48:34 -06:00
James Betker
119f17c808 Add testing capabilities for segformer & contrastive feature 2021-04-27 09:59:50 -06:00
James Betker
9bbe6fc81e Get segformer to a trainable state 2021-04-25 11:45:20 -06:00
James Betker
23e01314d4 Add dataset, ui for labeling and evaluator for pointwise classification 2021-04-23 17:17:13 -06:00
James Betker
fc623d4b5a Add segformer model. Start work on BYOL adaptation that will support training it. 2021-04-23 17:16:46 -06:00
James Betker
17555e7d07 misc adjustments for stylegan 2021-04-21 18:14:17 -06:00
James Betker
b687ef4cd0 Misc 2021-04-21 18:09:46 -06:00
James Betker
94e069bced Misc changes 2021-03-13 10:45:26 -07:00
James Betker
9fc3df3f5b Switched conv: add conversion function with allowlist 2021-03-13 10:44:56 -07:00
James Betker
cf9a6da889 Fix some bugs, checkin work on vqvae3 2021-03-02 20:56:19 -07:00
James Betker
f89ea5f1c6 Mods to support lightweight_gan model 2021-03-02 20:51:48 -07:00
James Betker
543d459b4e extract_temporal_squares script 
For extracting related patches across a video
 2021-02-08 08:10:24 -07:00
James Betker
39fd755baa New benchmark numbers 2021-02-08 08:09:41 -07:00
James Betker
784b96c059 Misc options to add support for training stylegan2-rosinality models: 
- Allow image_folder_dataset to normalize inbound images
- ExtensibleTrainer can denormalize images on the output path
- Support .webp - an output from LSUN
- Support logistic GAN divergence loss
- Support stylegan2 TF weight extraction for discriminator
- New injector that produces latent noise (with separated paths)
- Modify FID evaluator to be operable with rosinality-style GANs
 2021-02-08 08:09:21 -07:00
James Betker
e7be4bdff3 Revert 2021-02-05 08:43:07 -07:00
James Betker
6dec1f5968 Back to groupnorm 2021-02-05 08:42:11 -07:00
James Betker
336f807c8e lambda2 2021-02-05 00:00:24 -07:00
James Betker
025a5867c4 Use syncbatchnorm instead 2021-02-04 22:26:36 -07:00
James Betker
bb79fafb89 Fix groupnorm specification 2021-02-04 22:15:38 -07:00
James Betker
43da1f9c4b Convert lambda coupler to use groupnorm instead of batchnorm 2021-02-04 21:59:44 -07:00
James Betker
7070142805 Make vqvae3_hard more configurable 2021-02-04 09:03:22 -07:00
James Betker
b980028ca8 Add get_debug_values for vqvae_3_hardswitch 2021-02-03 14:12:24 -07:00
James Betker
1405ff06b8 Fix SwitchedConvHardRoutingFunction for current cuda router 2021-02-03 14:11:55 -07:00
James Betker
d7bec392dd ... 2021-02-02 23:50:25 -07:00
James Betker
b0a8fa00bc Visual dbg in vqvae3hs 2021-02-02 23:50:01 -07:00
James Betker
f5f91850fd hardswitch variant of vqvae3 2021-02-02 21:00:04 -07:00
James Betker
320edbaa3c Move switched_conv logic around a bit 2021-02-02 20:41:24 -07:00
James Betker
0dca36946f Hard Routing mods 
- Turns out my custom convolution was RIDDLED with backwards bugs, which is
   why the existing implementation wasn't working so well.
- Implements the switch logic from both Mixture of Experts and Switch Transformers
  for testing purposes.
 2021-02-02 20:35:58 -07:00
James Betker
29c1c3bede Register vqvae3 2021-01-29 15:26:28 -07:00
James Betker
bc20b4739e vqvae3 
Changes VQVAE as so:
- Reverts back to smaller codebook
- Adds an additional conv layer at the highest resolution for both the encoder & decoder
- Uses LeakyReLU on trunk
 2021-01-29 15:24:26 -07:00
James Betker
96bc80313c Add switch norm, up dropout rate, detach selector 2021-01-26 09:31:53 -07:00
James Betker
97d895aebe Add SrPixLoss, which focuses pixel-based losses on high-frequency regions 
of the image.
 2021-01-25 08:26:14 -07:00
James Betker
2cdac6bd09 Add PWCNet for human optical flow 2021-01-25 08:25:44 -07:00
James Betker
51b63b2aa6 Add switched_conv with hard routing and make vqvae use it. 2021-01-25 08:25:29 -07:00
James Betker
ae4ff4a1e7 Enable lambda visualization 2021-01-23 15:53:27 -07:00
James Betker
10ec6bda1d lambda nets in switched_conv and a vqvae to use it 2021-01-23 14:57:57 -07:00
James Betker
b374dcdd46 update vqvae to double codebook size for bottom quantizer 2021-01-23 13:47:07 -07:00
James Betker
dac7d768fa test uresnet playground mods 2021-01-23 13:46:43 -07:00
James Betker
1b8a26db93 New switched_conv 2021-01-23 13:46:30 -07:00
James Betker
557cdec116 misc 2021-01-23 13:45:17 -07:00
James Betker
d919ae7148 Add VQVAE with no Conv2dTranspose 2021-01-18 08:49:59 -07:00
James Betker
587a4f4050 resnet_unet_3 
I'm being really lazy here - these nets are not really different from each other
except at which layer they terminate. This one terminates at 2x downsampling,
which is simply indicative of a direction I want to go for testing these pixpro networks.
 2021-01-15 14:51:03 -07:00
James Betker
038b8654b6 Pixpro: unwrap losses 2021-01-13 11:54:25 -07:00
James Betker
8990801a3f Fix pixpro stochastic sampling bugs 2021-01-13 11:34:24 -07:00
James Betker
19475a072f Pixpro: Rather than using a latent square for pixpro, use an entirely stochastic sampling of the pixels 2021-01-13 11:26:51 -07:00
James Betker
d1007ccfe7 Adjustments to pixpro to allow training against networks with arbitrarily large structural latents 
- The pixpro latent now rescales the latent space instead of using a "coordinate vector", which
   **might** have performance implications.
- The latent against which the pixel loss is computed can now be a small, randomly sampled patch
   out of the entire latent, allowing further memory/computational discounts. Since the loss
   computation does not have a receptive field, this should not alter the loss.
- The instance projection size can now be separate from the pixel projection size.
- PixContrast removed entirely.
- ResUnet with full resolution added.
 2021-01-12 09:17:45 -07:00
James Betker
34f8c8641f Support training imagenet classifier 2021-01-11 20:09:16 -07:00
James Betker
f3db381fa1 Allow uresnet to use pretrained resnet50 2021-01-10 12:57:31 -07:00
James Betker
4119cd6240 Fix to image_folder_dataset to accomodate images with mismatched dimensions 2021-01-10 12:57:21 -07:00
James Betker
48f0d8964b Allow dist_backend to be specified in options 2021-01-09 20:54:32 -07:00
James Betker
14a868e8e6 byol playground updates 2021-01-09 20:54:21 -07:00
James Betker
7c6c7a8014 Fix process_video 2021-01-09 20:53:46 -07:00
James Betker
07168ecfb4 Enable vqvae to use a switched_conv variant 2021-01-09 20:53:14 -07:00
James Betker
41b7d50944 Update extract_square_images 2021-01-08 13:16:34 -07:00
James Betker
5a8156026a Did anyone ask for k-means clustering? 
This is so cool...
 2021-01-07 22:37:41 -07:00
James Betker
acf1535b14 Fix for randomresizedcrop injector 2021-01-07 16:31:43 -07:00
James Betker
659814c20f BYOL script updates 2021-01-07 16:31:28 -07:00
James Betker
de10c7246a Add injected noise into bypass maps 2021-01-07 16:31:12 -07:00
James Betker
04961b91cf Add random-crop injector 2021-01-07 12:14:55 -07:00
James Betker
61a86a3c1e VQVAE 2021-01-07 10:20:15 -07:00
James Betker
01a589e712 Adjustments to pixpro & resnet-unet 
I'm not really satisfied with what I got out of these networks on round 1.
Lets try again..
 2021-01-06 15:00:46 -07:00
James Betker
9680294430 Move byol scripts around 2021-01-06 14:52:17 -07:00
James Betker
2f2f87bbea Styled SR fixes 2021-01-05 20:14:39 -07:00
James Betker
9fed90393f Add lucidrains pixpro trainer 2021-01-05 20:14:22 -07:00
James Betker
39a94c74b5 Allow BYOL resnet playground to produce a latent dict 2021-01-04 20:11:29 -07:00
James Betker
ade2732c82 Transfer learning for styleSR 
This is a concept from "Lifelong Learning GAN", although I'm skeptical of it's novelty -
basically you scale and shift the weights for the generator and discriminator of a pretrained
GAN to "shift" into new modalities, e.g. faces->birds or whatever. There are some interesting
applications of this that I would like to try out.
 2021-01-04 20:10:48 -07:00
James Betker
2c65b6b28e More mods to support styledsr 2021-01-04 11:32:28 -07:00
James Betker
2225fe6ac2 Undo lucidrains changes for new discriminator 
This "new" code will live in the styledsr directory from now on.
 2021-01-04 10:57:09 -07:00
James Betker
40ec71da81 Move styled_sr into its own folder 2021-01-04 10:54:34 -07:00
James Betker
5916f5f7d4 Misc fixes 2021-01-04 10:53:53 -07:00
James Betker
4d8064c32c Modifications to allow partially trained stylegan discriminators to be used 2021-01-03 16:37:18 -07:00
James Betker
5e7ade0114 ImageFolderDataset - corrupt lq images alongside each other 2021-01-03 16:36:38 -07:00
James Betker
ce6524184c Do the last commit but in a better way 2021-01-02 22:24:12 -07:00
James Betker
edf9c38198 Make ExtensibleTrainer set the starting step for the LR scheduler 2021-01-02 22:22:34 -07:00
James Betker
bdbab65082 Allow optimizers to train separate param groups, add higher dimensional VGG discriminator 
Did this to support training 512x512px networks off of a pretrained 256x256 network.
 2021-01-02 15:10:06 -07:00
James Betker
193cdc6636 Move discriminators to the create_model paradigm 
Also cleans up a lot of old discriminator models that I have no intention
of using again.
 2021-01-01 15:56:09 -07:00
James Betker
7976a5825d srfid is incorrectly labeled 2021-01-01 13:00:59 -07:00
James Betker
f39179e85a styled_sr: fix bug when using initial_stride 2021-01-01 12:13:21 -07:00
James Betker
913fc3b75e Need init to pick up styled_sr 2021-01-01 12:10:32 -07:00
James Betker
aae65e6ed8 Mods to byol_resnet_playground for large batches 2021-01-01 11:59:54 -07:00
James Betker
e992e18767 Add initial_stride term to style_sr 
Also fix fid and a networks.py issue.
 2021-01-01 11:59:36 -07:00
James Betker
9864fe4c04 Fix for train.py 2021-01-01 11:59:00 -07:00
James Betker
e214e6ce33 Styled SR model 2020-12-31 20:54:18 -07:00
James Betker
0eb1f4dd67 Revert "Get rid of CUDA_VISIBLE_DEVICES" 
It is actually necessary for training in distributed mode. Only
do it then.
 2020-12-31 10:31:40 -07:00
James Betker
8de5a02a48 byol_resnet_playground 
Similar to the spinenet playground, but tinkers with resnet instead
 2020-12-31 10:15:04 -07:00
James Betker
8f18b2709e Get rid of CUDA_VISIBLE_DEVICES 
It is not clear to me what the purpose of this is, but it has recently
started causing failures.
 2020-12-31 10:13:58 -07:00
James Betker
1de1fa30ac Disable refs and centers altogether in single_image_dataset 
I suspect that this might be a cause of failures on parallel datasets.
Plus it is unnecessary computation.
 2020-12-31 10:13:24 -07:00
James Betker
8f0984cacf Add sr_fid evaluator 2020-12-30 20:18:58 -07:00
James Betker
b1fb82476b Add gp debug (fix) 2020-12-30 15:26:54 -07:00
James Betker
9c53314ea2 Add gradient penalty visual debug 2020-12-30 09:51:59 -07:00
James Betker
63cf3d3126 Injector auto-registration 
I love it!
 2020-12-29 20:58:02 -07:00
James Betker
a777c1e4f9 Misc script fixes 2020-12-29 20:25:09 -07:00
James Betker
9dc3c8f0ff Script updates 2020-12-29 20:24:41 -07:00
James Betker
ba543d1152 Glean mods 
- Fixes fixed upscale factor issues
- Refines a few ops to decrease computation & parameterization
 2020-12-27 12:25:06 -07:00
James Betker
5e2e605a50 Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-12-26 13:51:19 -07:00
James Betker
f9be049adb GLEAN mod to support custom initial strides 2020-12-26 13:51:14 -07:00
James Betker
2706a84f15 Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-12-26 13:50:34 -07:00
James Betker
90e2362c00 Fix bug with full_image_dataset 2020-12-26 13:50:27 -07:00
James Betker
3fd627fc62 Mods to support image classification & filtering 2020-12-26 13:49:27 -07:00
James Betker
10fdfa1563 Migrate generators to dynamic model registration 2020-12-24 23:02:10 -07:00
James Betker
29db7c7a02 Further mods to BYOL 2020-12-24 09:28:41 -07:00
James Betker
036684893e Add LARS optimizer & support for BYOL idiosyncrasies 
- Added LARS and SGD optimizer variants that support turning off certain
  features for BN and bias layers
- Added a variant of pytorch's resnet model that supports gradient checkpointing.
- Modify the trainer infrastructure to support above
- Fix bug with BYOL (should have been nonfunctional)
 2020-12-23 20:33:43 -07:00
James Betker
1bbcb96ee8 Implement a few changes to support training BYOL networks 2020-12-23 10:50:23 -07:00
James Betker
2437b33e74 Fix srflow_latent_space_playground bug 2020-12-22 15:42:38 -07:00
James Betker
e7aeb17404 ImageFolder dataset: allow intermediary downscale before corrupt 
For massive upscales (ex: 8x), corruption does almost nothing when applied
at the HQ level. This patch adds support to perform corruption at a specified
intermediary scale. The dataset downscales to this level, performs the corruption,
then downscales the rest of the way to get the LQ image.
 2020-12-22 15:42:21 -07:00
James Betker
7938f9f50b Fix bug with single_image_dataset which prevented working on multiple directories from working 2020-12-19 15:13:46 -07:00
James Betker
ae666dc520 Fix bugs with srflow after refactor 2020-12-19 10:28:23 -07:00
James Betker
4328c2f713 Change default ReLU slope to .2 BREAKS COMPATIBILITY 
This conforms my ConvGnLelu implementation with the generally accepted negative_slope=.2. I have no idea where I got .1. This will break backwards compatibility with some older models but will likely improve their performance when freshly trained. I did some auditing to find what these models might be, and I am not actively using any of them, so probably OK.
 2020-12-19 08:28:03 -07:00
James Betker
9377d34ac3 glean mods 2020-12-19 08:26:07 -07:00
James Betker
f35c034fa5 Add trainer readme 2020-12-18 16:52:16 -07:00
James Betker
e82f4552db Update other docs with dumb config options 2020-12-18 16:21:28 -07:00
James Betker
92f9a129f7 GLEAN! 2020-12-18 16:04:19 -07:00
James Betker
c717765bcb Notes for lucidrains converter. 2020-12-18 09:55:38 -07:00
James Betker
b4720ea377 Move stylegan to new location 2020-12-18 09:52:36 -07:00
James Betker
1708136b55 Commit my attempt at "conforming" the lucidrains stylegan implementation to the reference spec. Not working. will probably be abandoned. 2020-12-18 09:51:48 -07:00
James Betker
209332292a Rosinality stylegan fix 2020-12-18 09:50:41 -07:00
James Betker
d875ca8342 More refactor changes 2020-12-18 09:24:31 -07:00
James Betker
5640e4efe4 More refactoring 2020-12-18 09:18:34 -07:00
James Betker
b905b108da Large cleanup 
Removed a lot of old code that I won't be touching again. Refactored some
code elements into more logical places.
 2020-12-18 09:10:44 -07:00
James Betker
2f0a52b7db misc changes 2020-12-18 08:53:45 -07:00
James Betker
a8179ff53c Image label work 2020-12-18 08:53:18 -07:00
James Betker
3074f41877 Get rosinality model converter to work 
Mostly, just needed to remove the custom cuda ops, not so bueno on Windows.
 2020-12-17 16:03:39 -07:00
James Betker
e838c6e75b Rosinality stylegan2 port 2020-12-17 14:18:46 -07:00
James Betker
12cf052889 Add an image patch labeling UI 2020-12-17 10:16:21 -07:00
James Betker
49327b99fe SRFlow outputs RRDB output 2020-12-16 10:28:02 -07:00
James Betker
c25b49bb12 Clean up of SRFlowNet_arch 2020-12-16 10:27:38 -07:00
James Betker
42ac8e3eeb Remove unnecessary comment from SRFlowNet 2020-12-16 09:43:07 -07:00
James Betker
fb2cfc795b Update requirements, add image_patch_classifier tool 2020-12-16 09:42:50 -07:00
James Betker
09de3052ac Add softmax to spinenet classification head 2020-12-16 09:42:15 -07:00
James Betker
4310e66848 Fix bug in 'corrupt_before_downsize=true' 2020-12-16 09:41:59 -07:00
James Betker
8661207d57 Merge branch 'gan_lab' of https://github.com/neonbjb/DL-Art-School into gan_lab 2020-12-15 17:16:48 -07:00
James Betker
fc376d34b2 Spinenet with logits head 2020-12-15 17:16:19 -07:00
James Betker
8e0e883050 Mods to support labeled datasets & random augs for those datasets 2020-12-15 17:15:56 -07:00
James Betker
e5a3e6b9b5 srflow latent space misc 2020-12-14 23:59:49 -07:00
James Betker
1e14635d88 Add exclusions to extract_subimages_with_ref 2020-12-14 23:59:41 -07:00
James Betker
0a19e53df0 BYOL mods 2020-12-14 23:59:11 -07:00
James Betker
ef7eabf457 Allow RRDB to upscale 8x 2020-12-14 23:58:52 -07:00
James Betker
087e9280ed Add labeling feature to image_folder_dataset 2020-12-14 23:58:37 -07:00
James Betker
ec0ee25f4b Structural latents checkpoint 2020-12-11 12:01:09 -07:00
James Betker
26ceca68c0 BYOL with structure! 2020-12-10 15:07:35 -07:00
James Betker
9c5e272a22 Script to extract models from a wrapped BYOL model 2020-12-10 09:57:52 -07:00
James Betker
a5630d282f Get rid of 2nd trainer 2020-12-10 09:57:38 -07:00
James Betker
8e4b9f42fd New BYOL dataset which uses a form of RandomCrop that lends itself to 
structural guidance to the latents.
 2020-12-10 09:57:18 -07:00
James Betker
c203cee31e Allow swapping to torch DDP as needed in code 2020-12-09 15:03:59 -07:00
James Betker
66cbae8731 Add random_dataset for testing 2020-12-09 14:55:05 -07:00
James Betker
97ff25a086 BYOL! 
Man, is there anything ExtensibleTrainer can't train? :)
 2020-12-08 13:07:53 -07:00
James Betker
5369cba8ed Stage 2020-12-08 00:33:07 -07:00
James Betker
bca59ed98a Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-12-07 12:51:04 -07:00
James Betker
ea56eb61f0 Fix DDP errors for discriminator 
- Don't define training_net in define_optimizers - this drops the shell and leads to problems downstream
- Get rid of support for multiple training nets per opt. This was half baked and needs a better solution if needed
   downstream.
 2020-12-07 12:50:57 -07:00
James Betker
c0aeaabc31 Spinenet playground 2020-12-07 12:49:32 -07:00
James Betker
88fc049c8d spinenet latent playground! 2020-12-05 20:30:36 -07:00
James Betker
11155aead4 Directly use dataset keys 
This has been a long time coming. Cleans up messy "GT" nomenclature and simplifies ExtensibleTraner.feed_data
 2020-12-04 20:14:53 -07:00
James Betker
8a83b1c716 Go back to apex DDP, fix distributed bugs 2020-12-04 16:39:21 -07:00
James Betker
7a81d4e2f4 Revert gaussian loss changes 2020-12-04 12:49:20 -07:00
James Betker
711780126e Cleanup 2020-12-03 23:42:51 -07:00
James Betker
ac7256d4a3 Do tqdm reporting when calculating flow_gaussian_nll 2020-12-03 23:42:29 -07:00
James Betker
dc9ff8e05b Allow the majority of the srflow steps to checkpoint 2020-12-03 23:41:57 -07:00
James Betker
06d1c62c5a iGPT support! 
Sweeeeet
 2020-12-03 15:32:21 -07:00
James Betker
c18adbd606 Delete mdcn & panet 
Garbage, all of it.
 2020-12-02 22:25:57 -07:00
James Betker
f2880b33c9 Get rid of mean shift from MDCN 2020-12-02 14:18:33 -07:00
James Betker
8a00f15746 Implement FlowGaussianNll evaluator 2020-12-02 14:09:54 -07:00
James Betker
edf408508c Fix discriminator 2020-12-01 17:45:56 -07:00
James Betker
c963e5f2ce Add ImageFolderDataset 
This one has been a long time coming.. How does torch not have something like this?
 2020-12-01 17:45:37 -07:00
James Betker
9a421a41f4 SRFlow: accomodate mismatches between global scale and flow_scale 2020-12-01 11:11:51 -07:00
James Betker
8f65f81ddb Adjustments to subimage extractor 2020-12-01 11:11:30 -07:00
James Betker
e343722d37 Add stepped rrdb 2020-12-01 11:11:15 -07:00
James Betker
2e0bbda640 Remove unused archs 2020-12-01 11:10:48 -07:00
James Betker
a1c8300052 Add mdcn 2020-11-30 16:14:21 -07:00
James Betker
1e0f69e34b extra_conv in gn discriminator, multiframe support in rrdb. 2020-11-29 15:39:50 -07:00
James Betker
da604752e6 Misc RRDB changes 2020-11-29 12:21:31 -07:00
James Betker
f2422f1d75 Latent space playground 2020-11-29 09:33:29 -07:00
James Betker
a1d4c9f83c multires rrdb work 2020-11-28 14:35:46 -07:00
James Betker
929cd45c05 Fix for RRDB scale 2020-11-27 21:37:10 -07:00
James Betker
71fa532356 Adjustments to how flow networks set size and scale 2020-11-27 21:37:00 -07:00
James Betker
6f958bb150 Maybe this is necessary after all? 2020-11-27 15:21:13 -07:00
James Betker
ef8d5f88c1 Bring split gaussian nll out of split so it can be computed accurately with the rest of the nll component 2020-11-27 13:30:21 -07:00
James Betker
11d2b70bdd Latent space playground work 2020-11-27 12:03:16 -07:00
James Betker
4ab49b0d69 RRDB disc work 2020-11-27 12:03:08 -07:00
James Betker
6de4dabb73 Remove srflow (modified version) 
Starting from orig and re-working from there.
 2020-11-27 12:02:06 -07:00
James Betker
5f5420ff4a Update to srflow_latent_space_playground 2020-11-26 20:31:21 -07:00
James Betker
fd356580c0 Play with lambdas 2020-11-26 20:30:55 -07:00
James Betker
0c6d7971b9 Dataset documentation 2020-11-26 11:58:39 -07:00
James Betker
45a489110f Fix datasets 2020-11-26 11:50:38 -07:00
James Betker
5edaf085e0 Adjustments to latent_space_playground 2020-11-25 15:52:36 -07:00
James Betker
205c9a5335 Learn how to functionally use srflow networks 2020-11-25 13:59:06 -07:00
James Betker
cb045121b3 Expose srflow rrdb 2020-11-24 13:20:20 -07:00
James Betker
f3c1fc1bcd Dataset modifications 2020-11-24 13:20:12 -07:00
James Betker
f6098155cd Mods to tecogan to allow use of embeddings as input 2020-11-24 09:24:02 -07:00
James Betker
b10bcf6436 Rework stylegan_for_sr to incorporate structure as an adain block 2020-11-23 11:31:11 -07:00
James Betker
519ba6f10c Support 2x RRDB with 4x srflow 2020-11-21 14:46:15 -07:00
James Betker
cad92bada8 Report logp and logdet for srflow 2020-11-21 10:13:05 -07:00
James Betker
c37d3faa58 More adjustments to srflow_orig 2020-11-20 19:38:33 -07:00
James Betker
d51d12a41a Adjustments to srflow to (maybe?) fix training 2020-11-20 14:44:24 -07:00
James Betker
6c8c35ac47 Support training RRDB encoder [srflow] 2020-11-20 10:03:06 -07:00
James Betker
5ccdbcefe3 srflow_orig integration 2020-11-19 23:47:24 -07:00
James Betker
f80acfcab6 Throw if dataset isn't going to work with force_multiple setting 2020-11-19 23:47:00 -07:00
James Betker
2b2d754d8e Bring in an original SRFlow implementation for reference 2020-11-19 21:42:39 -07:00
James Betker
1e0d7be3ce "Clean up" SRFlow 2020-11-19 21:42:24 -07:00
James Betker
d7877d0a36 Fixes to teco losses and translational losses 2020-11-19 11:35:05 -07:00
James Betker
b2a05465fc Fix missing requirements 2020-11-18 10:16:39 -07:00
James Betker
5c10264538 Remove pyramid_disc hard dependencies 2020-11-17 18:34:11 -07:00
James Betker
6b679e2b51 Make grad_penalty available to classical discs 2020-11-17 18:31:40 -07:00
James Betker
8a19c9ae15 Add additive mode to rrdb 2020-11-16 20:45:09 -07:00
James Betker
2a507987df Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-11-15 16:16:30 -07:00
James Betker
931ed903c1 Allow combined additive loss 2020-11-15 16:16:18 -07:00
James Betker
4b68116977 import fix 2020-11-15 16:15:42 -07:00
James Betker
98eada1e4c More circular dependency fixes + unet fixes 2020-11-15 11:53:35 -07:00
James Betker
e587d549f7 Fix circular imports 2020-11-15 11:32:35 -07:00
James Betker
99f0cfaab5 Rework stylegan2 divergence losses 
Notably: include unet loss
 2020-11-15 11:26:44 -07:00
James Betker
ea94b93a37 Fixes for unet 2020-11-15 10:38:33 -07:00
James Betker
89f56b2091 Fix another import 2020-11-14 22:10:45 -07:00
James Betker
9af049c671 Import fix for unet 2020-11-14 22:09:18 -07:00
James Betker
5cade6b874 Move stylegan2 around, bring in unet 2020-11-14 22:04:48 -07:00
James Betker
4c6b14a3f8 Allow extract_square_images to work on multiple images 2020-11-14 20:24:05 -07:00
James Betker
125cb16dce Add a FID evaluator for stylegan with structural guidance 2020-11-14 20:16:07 -07:00
James Betker
c9258e2da3 Alter how structural guidance is given to stylegan 2020-11-14 20:15:48 -07:00
James Betker
3397c83447 Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-11-14 09:30:09 -07:00
James Betker
423ee7cb90 Allow attention to be specified for stylegan2 2020-11-14 09:29:53 -07:00
James Betker
ec621c69b5 Fix train bug 2020-11-14 09:29:08 -07:00
James Betker
cdc5ac30e9 oddity 2020-11-13 20:11:57 -07:00
James Betker
f406a5dd4c Mods to support stylegan2 in SR mode 2020-11-13 20:11:50 -07:00
James Betker
9c3d0b7560 Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-11-13 20:10:47 -07:00
James Betker
67bf55495b Allow hq_batched_key to be specified 2020-11-13 20:10:12 -07:00
James Betker
0b96811611 Fix another issue with gpu ids getting thrown all over hte place 2020-11-13 20:05:52 -07:00
James Betker
c47925ae34 New image extractor utility 2020-11-13 11:04:03 -07:00
James Betker
a07e1a7292 Add separate Evaluator module and FID evaluator 2020-11-13 11:03:54 -07:00
James Betker
080ad61be4 Add option to work with nonrandom latents 2020-11-12 21:23:50 -07:00
James Betker
566b99ca75 GP adjustments for stylegan2 2020-11-12 16:44:51 -07:00
James Betker
fc55bdb24e Mods to how wandb are integrated 2020-11-12 15:45:25 -07:00
James Betker
44a19cd37c ExtensibleTrainer mods to support advanced checkpointing for stylegan2 
Basically: stylegan2 makes use of gradient-based normalizers. These
make it so that I cannot use gradient checkpointing. But I love gradient
checkpointing. It makes things really, really fast and memory conscious.

So - only don't checkpoint when we run the regularizer loss. This is a
bit messy, but speeds up training by at least 20%.

Also: pytorch: please make checkpointing a first class citizen.
 2020-11-12 15:45:07 -07:00
James Betker
db9e9e28a0 Fix an issue where GPU0 was always being used in non-ddp 
Frankly, I don't understand how this has ever worked. WTF.
 2020-11-12 15:43:01 -07:00
James Betker
2d3449d7a5 stylegan2 in ml art school! 2020-11-12 15:42:05 -07:00
James Betker
fd97573085 Fixes 2020-11-11 21:49:06 -07:00
James Betker
88f349bdf1 Enable usage of wandb 2020-11-11 21:48:56 -07:00
James Betker
1c065c41b4 Revert "..." 
This reverts commit 4b92191880.
 2020-11-11 17:24:27 -07:00
James Betker
4b92191880 ... 2020-11-11 14:12:40 -07:00
James Betker
12b57bbd03 Add residual blocks to pyramid disc 2020-11-11 13:56:45 -07:00
James Betker
b4136d766a Back to pyramids, no rrdb 2020-11-11 13:40:24 -07:00
James Betker
42a97de756 Convert PyramidRRDBDisc to RRDBDisc 
Had numeric stability issues. This probably makes more sense anyways.
 2020-11-11 12:14:14 -07:00
James Betker
72762f200c PyramidRRDB net 2020-11-11 11:25:49 -07:00
James Betker
a1760f8969 Adapt srg2 for video 2020-11-10 16:16:41 -07:00
James Betker
b742d1e5a5 When skipping steps via "every", still run nontrainable injection points 2020-11-10 16:09:17 -07:00
James Betker
91d27372e4 rrdb with adain latent 2020-11-10 16:08:54 -07:00
James Betker
6a2fd5f7d0 Lots of new discriminator nets 2020-11-10 16:06:54 -07:00
James Betker
4e5ba61ae7 SRG2classic further re-integration 2020-11-10 16:06:14 -07:00
James Betker
9e2c96ad5d More latent work 2020-11-07 20:38:56 -07:00
James Betker
6be6c92e5d Fix yet ANOTHER OBO error in multi_frame_dataset 2020-11-06 20:38:34 -07:00
James Betker
0cf52ef52c latent work 2020-11-06 20:38:23 -07:00
James Betker
34d319585c Add srflow arch 2020-11-06 20:38:04 -07:00
James Betker
4469d2e661 More work on RRDB with latent 2020-11-05 22:13:05 -07:00
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
c21088e238 Fix OBO error in multi_frame_dataset 
In some datasets, this meant one frame was included in a sequence where it didn't belong. In datasets with mismatched chunk sizes, this resulted in an error.
 2020-11-03 14:32:06 -07:00
James Betker
e990be0449 Improve ignore_first logic 2020-11-03 11:56:32 -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
85c545835c Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-11-02 08:48:15 -07:00
James Betker
f13fdd43ed Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-11-02 08:47:42 -07:00
James Betker
fed16abc22 Report chunking errors 2020-11-02 08:47:18 -07:00
James Betker
a51daacde2 Fix reporting of d_fake_diff for generators 2020-11-02 08:45:46 -07:00
James Betker
3676f26d94 Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-10-31 20:55:45 -06:00
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
ea8c20c0e2 Fix bug with multiscale_dataset 2020-10-31 20:54:41 -06:00
James Betker
bb39d3efe5 Bump image corruption factor a bit 2020-10-31 20:50:24 -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
b24ff3c88d Fix bug that causes multiscale dataset to crash 2020-10-30 14:01:24 -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
25b007a0f5 Increase jpeg corruption & add error 2020-10-28 17:37:39 -06:00
James Betker
796659b0ac Add 'jpeg-normal' corruption 2020-10-28 16:40:47 -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
31cf1ac98d Retrofit full_image_dataset to work with new arch. 2020-10-27 10:26:19 -06:00
James Betker
ade0a129da Include psnr in test.py 2020-10-27 10:25:42 -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
b2f803588b Fix multi_modal_train.py 2020-10-26 11:10:22 -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
327cdbe110 Support configurable multi-modal training 2020-10-24 11:57:39 -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
8e5b6682bf Add PairedFrameDataset 2020-10-23 20:58:07 -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
8636492db0 Copy train.py mods to train2 2020-10-22 17:16:36 -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
f9dc472f63 Misc nonfunctional mods to datasets 2020-10-22 10:16:17 -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
b54de69153 Misc 2020-10-21 11:08:21 -06:00
James Betker
71c3820d2d Fix process_video 2020-10-21 11:08:12 -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
d8c6a4bbb8 Misc 2020-10-20 12:56:52 -06:00
James Betker
aba83e7497 Don't apply jpeg corruption & noise corruption together 
This causes some severe noise.
 2020-10-20 12:56:35 -06:00
James Betker
111450f4e7 Use areal interpolate for multiscale_dataset 2020-10-19 15:30:25 -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
331c40f0c8 Allow starting step to be forced 
Useful for testing purposes or to force a validation.
 2020-10-19 15:23:04 -06:00
James Betker
8ca566b621 Revert "Misc" 
This reverts commit 0e3ea63a14.

# Conflicts:
#	codes/test.py
#	codes/train.py
 2020-10-19 13:34:54 -06:00
James Betker
b28e4d9cc7 Add spread loss 
Experimental loss that peaks around 0.
 2020-10-19 11:31:19 -06:00
James Betker
9b9a6e5925 Add get_paths() to base_unsupervised_image_dataset 2020-10-19 11:30:06 -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
ffad0e0422 Allow image corruption in multiscale dataset 2020-10-19 10:10:27 -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
c7f3fc4dd9 Enable chunk_with_reference to work without centers 
Moving away from this so it doesn't matter too much. Also fixes an issue
with the "ignore" flag.
 2020-10-17 20:09:08 -06:00
James Betker
b45e132a9d Allow first n tiles to be ignored 
Helps zoom in with chunked dataset
 2020-10-17 09:45:03 -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
96f1be30ed Add use_generator_as_filter 2020-10-16 20:43:55 -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
1ba01d69b5 Move datasets to INTER_AREA interpolation for downsizing 
Looks **FAR** better visually
 2020-10-15 17:18:23 -06:00
James Betker
d56745b2ec JPEG-broad adjustment 2020-10-15 10:14:51 -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
1dc0b05428 Add multiscale dataset 2020-10-15 10:12:50 -06:00
James Betker
0f4e03183f New image corruptor gradations 2020-10-15 10:12:25 -06:00
James Betker
1f20d59c31 Revert big switch back 2020-10-14 11:03:34 -06:00
James Betker
9815980329 Update SwitchedConv 2020-10-13 20:57:12 -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
c1a00f31b7 Update switched_conv 2020-10-12 10:37:45 -06:00
James Betker
d7d7590f3e Fix constant injector - wasn't working in test 2020-10-12 10:36:30 -06:00
James Betker
e7cf337dba Fix bug with chunk_with_reference 2020-10-12 10:23:03 -06:00
James Betker
ce163ad4a9 Update SSGdeep 2020-10-12 10:22:08 -06:00
James Betker
2bc5701b10 misc 2020-10-12 10:21:25 -06:00
James Betker
3409d88a1c Add PANet arch 2020-10-12 10:20:55 -06:00
James Betker
7cbf4fa665 Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-10-11 08:33:30 -06:00
James Betker
92cb83958a Return zeros rather than None when image cant be read 2020-10-11 08:33:18 -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
61e5047c60 Fix loss accumulator when buffers are not filled 
They were reporting incorrect losses.
 2020-10-09 19:47:59 -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
b3d0baaf17 Improve multiframe dataset memory usage 2020-10-09 08:40:00 -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
3cc56cd00b Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-10-08 16:12:05 -06:00
James Betker
7d8d9dafbb misc 2020-10-08 16:12:00 -06:00
James Betker
856ef4d21d Update switched_conv 2020-10-08 16:10:23 -06:00
James Betker
1eb516d686 Fix more distributed bugs 2020-10-08 14:32:45 -06:00
James Betker
b36ba0460c Fix multi-frame dataset OBO error 2020-10-08 12:21:04 -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
29bf78d791 Update switched_conv submodule 2020-10-07 23:11:50 -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
a62a5dbb5f Clone and detach in recursively_detach 2020-10-07 12:41:00 -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
b2c4b2a16d Move gpu_ids out of if statement 2020-10-06 20:40:20 -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
4290918359 Add distributed_checkpoint for more efficient checkpoints 2020-10-06 20:38:38 -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
0e3ea63a14 Misc 2020-10-05 18:01:50 -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
fc396baf1a Move loaded_options to util 
Doesn't seem to work with python 3.6
 2020-10-03 20:29:06 -06:00
James Betker
2d8e9a9d30 Options fix? 2020-10-03 20:27:12 -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
43c6c67fd1 Merge remote-tracking branch 'origin/gan_lab' into gan_lab 2020-10-03 16:17:31 -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
c896939523 Fix recursive checkpoint 2020-10-03 16:15:52 -06:00
James Betker
3cbb9ecd45 Misc 2020-10-03 16:15:42 -06:00
James Betker
35731502c3 Fix checkpoint recursion 2020-10-03 12:52:50 -06:00
James Betker
9b4ed82093 Get rid of unused convs in spsr7 2020-10-03 11:36:26 -06:00
James Betker
b2b81b13a4 Remove recursive utils import 2020-10-03 11:30:05 -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
21d3bb83b2 Use tqdm reporting with validation 2020-10-03 11:16:39 -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
efbf6b737b Update validate_data to work with SingleImageDataset 2020-10-02 08:58:34 -06:00
James Betker
35469f08e2 Spsr 8 2020-10-02 08:58:15 -06:00
James Betker
c9a9e5c525 Prompt user for gpu_id if multiple gpus are detected 2020-10-01 17:24:50 -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
57814f18cf More features for multi-frame-dataset 2020-09-28 14:26:15 -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
7dff802144 Add MultiFrameDataset 
Retrieves video sequence patches rather than single images.
 2020-09-27 11:13:06 -06:00
James Betker
d8c3fc9327 Fix random noise corruptor 
It was functioning as a color shift
 2020-09-27 11:12:24 -06:00
James Betker
c85da79697 Move many dataset functions into a base class 2020-09-27 11:11:58 -06:00
James Betker
eb12b5f887 Misc 2020-09-26 21:27:17 -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
254cb1e915 More dataset integration work 2020-09-25 22:19:38 -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
1cf73c2cce Fix dataset for a val set that includes lq 2020-09-24 18:01:07 -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
384e3d54cc Extract images into jpg, have a multiplier & size threshold 2020-09-21 12:36:03 -06:00
James Betker
bde35ced47 Fix recursive detach 2020-09-20 19:08:13 -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
dab8ab8a8f Offer option to configure the size of the normal distribution that the target size is drawn from 2020-09-20 11:59:31 -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
4f75cf0f02 Revert "Full image dataset operates on lists" 
Going with an entirely new dataset instead..

This reverts commit 36ec32bf11.
 2020-09-18 09:50:43 -06:00
James Betker
36ec32bf11 Full image dataset operates on lists 2020-09-18 09:50:26 -06:00
James Betker
3cb2a9a9d3 New dataset, initial work 2020-09-18 09:49:13 -06:00
James Betker
9a17ade550 Some convenience adjustments to ExtensibleTrainer 2020-09-17 21:05:32 -06:00
James Betker
57fc3f490c Add script for extracting image tiles with reference images 2020-09-17 13:30:51 -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
f5cd23e2d5 Further patch size adjustments 2020-09-16 16:50:35 -06:00
James Betker
723754c133 Update attention debugger outputting for SSG 2020-09-16 13:09:46 -06:00
James Betker
0b047e5f80 Increase scale of the patch selector random distribution 
This will cause larger slices of an image to appear more frequently,
increasing the difficulty of the generator.
 2020-09-16 08:27:42 -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
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
c833bd1eac Misc changes 2020-09-15 20:57:59 -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
94deab2792 Fix error serving gt_fullsize_ref in full_image_dataset 2020-09-14 15:05:44 -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
4c2ee66fe4 Fix video processor 2020-09-11 13:10:14 -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
c41dc9a48c Add missing requirements 2020-09-09 10:46:08 -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