VictorMylle/Kohya-ss-sd-scripts
1
0
Fork 1
mirror of https://github.com/kohya-ss/sd-scripts.git synced 2026-04-09 06:45:09 +00:00
Code Issues Packages Projects Releases Wiki Activity
2,057 Commits 25 Branches 38 Tags
c5b803ce94bd70812e6979ac7b986a769659b14e
Commit Graph

378 Commits

Author SHA1 Message Date
Kohya S
c5b803ce94 rng state management: Implement functions to get and set RNG states for consistent validation 2025-02-04 21:59:09 +09:00
Kohya S
45ec02b2a8 use same noise for every validation 2025-01-27 22:10:38 +09:00
Kohya S
0750859133 validation: Implement timestep-based validation processing 2025-01-27 21:56:59 +09:00
Kohya S
29f31d005f add network.train()/eval() for validation 2025-01-27 21:35:43 +09:00
Kohya S
b6a3093216 call optimizer eval/train fn before/after validation 2025-01-27 21:22:11 +09:00
Kohya S
86a2f3fd26 Fix gradient handling when Text Encoders are trained 2025-01-27 21:10:52 +09:00
Kohya S
532f5c58a6 formatting 2025-01-27 20:50:42 +09:00
rockerBOO
c04e5dfe92 Fix loss recorder on 0. Fix validation for cached runs. Assert on validation dataset 2025-01-23 09:57:24 -05:00
rockerBOO
25929dd0d7 Remove Validating... print to fix output layout 2025-01-12 15:38:57 -05:00
rockerBOO
ee9265cf26 Fix validate_every_n_steps for gradient accumulation 2025-01-12 14:56:35 -05:00
rockerBOO
0456858992 Fix validate_every_n_steps always running first step 2025-01-12 14:47:49 -05:00
rockerBOO
2bbb40ce51 Fix regularization images with validation 
Adding metadata recording for validation arguments
Add comments about the validation split for clarity of intention
 2025-01-12 14:29:50 -05:00
rockerBOO
4c61adc996 Add divergence to logs 
Divergence is the difference between training and validation to
allow a clear value to indicate the difference between the two
in the logs.
 2025-01-12 13:18:26 -05:00
rockerBOO
1e61392cf2 Revert bucket_reso_steps to correct 64 2025-01-08 18:43:26 -05:00
rockerBOO
556f3f1696 Fix documentation, remove unused function, fix bucket reso for sd1.5, fix multiple datasets 2025-01-08 13:41:15 -05:00
rockerBOO
1231f5114c Remove unused train_util code, fix accelerate.log for wandb, add init_trackers library code 2025-01-07 22:31:41 -05:00
rockerBOO
742bee9738 Set validation steps in multiple lines for readability 2025-01-06 17:34:23 -05:00
rockerBOO
fcb2ff010c Clean up some validation help documentation 2025-01-06 11:39:32 -05:00
rockerBOO
f8850296c8 Fix validate epoch, cleanup imports 2025-01-06 11:34:10 -05:00
rockerBOO
c64d1a22fc Add validate_every_n_epochs, change name validate_every_n_steps 2025-01-06 11:30:21 -05:00
rockerBOO
1c63e7cc49 Cleanup unused code and formatting 2025-01-06 11:07:47 -05:00
rockerBOO
bbf6bbd5ea Use self.get_noise_pred_and_target and drop fixed timesteps 2025-01-06 10:48:38 -05:00
rockerBOO
1c0ae306e5 Add missing functions for training batch 2025-01-03 15:43:02 -05:00
rockerBOO
1f9ba40b8b Add step break for validation epoch. Remove unused variable 2025-01-03 15:32:07 -05:00
rockerBOO
0522070d19 Fix training, validation split, revert to using upstream implemenation 2025-01-03 15:20:25 -05:00
rockerBOO
58bfa36d02 Add seed help clarifying info 2025-01-03 02:00:28 -05:00
rockerBOO
534059dea5 Typos and lingering is_train 2025-01-03 01:18:15 -05:00
rockerBOO
d23c7322ee Merge remote-tracking branch 'hina/feature/val-loss' into validation-loss-upstream 
Modified implementation for process_batch and cleanup validation
recording
 2025-01-03 00:48:08 -05:00
rockerBOO
7f6e124c7c Merge branch 'gesen2egee/val' into validation-loss-upstream 
Modified various implementations to restore original behavior
 2025-01-02 23:04:38 -05:00
gesen2egee
8743532963 val 2025-01-02 15:57:12 -05:00
Hina Chen
cb89e0284e Change val latent loss compare 2024-12-28 11:57:04 +08:00
Hina Chen
64bd5317dc Split val latents/batch and pick up val latents shape size which equal to training batch. 2024-12-28 11:42:15 +08:00
Hina Chen
62164e5792 Change val loss calculate method 2024-12-27 17:28:05 +08:00
Hina Chen
05bb9183fa Add Validation loss for LoRA training 2024-12-27 16:47:59 +08:00
Kohya S.
14c9ba925f Merge pull request #1811 from rockerBOO/schedule-free-prodigy 
Allow unknown schedule-free optimizers to continue to module loader
 2024-12-01 21:51:25 +09:00
Kohya S
cc11989755 fix: refactor huber-loss calculation in multiple training scripts 2024-12-01 21:20:28 +09:00
rockerBOO
6593cfbec1 Fix d * lr step log 2024-11-29 14:16:24 -05:00
rockerBOO
87f5224e2d Support d*lr for ProdigyPlus optimizer 2024-11-29 14:16:00 -05:00
recris
420a180d93 Implement pseudo Huber loss for Flux and SD3 2024-11-27 18:37:09 +00:00
Kohya S
2cb7a6db02 feat: add block swap for FLUX.1/SD3 LoRA training 2024-11-12 21:39:13 +09:00
Kohya S
cde90b8903 feat: implement block swapping for FLUX.1 LoRA (WIP) 2024-11-12 08:49:05 +09:00
kohya-ss
1065dd1b56 Fix to work dropout_rate for TEs 2024-10-27 19:36:36 +09:00
kohya-ss
a1255d637f Fix SD3 LoRA training to work (WIP) 2024-10-27 17:03:36 +09:00
Kohya S
db2b4d41b9 Add dropout rate arguments for CLIP-L, CLIP-G, and T5, fix Text Encoders LoRA not trained 2024-10-27 16:42:58 +09:00
kohya-ss
d2c549d7b2 support SD3 LoRA 2024-10-25 21:58:31 +09:00
Kohya S
5fba6f514a Merge branch 'dev' into sd3 2024-10-25 19:03:27 +09:00
catboxanon
e1b63c2249 Only add warning for deprecated scaling vpred loss function 2024-10-21 08:12:53 -04:00
catboxanon
8fc30f8205 Fix training for V-pred and ztSNR 
1) Updates debiased estimation loss function for V-pred.
2) Prevents now-deprecated scaling of loss if ztSNR is enabled.
 2024-10-21 07:34:33 -04:00
Kohya S
3cc5b8db99 Diff Output Preserv loss for SDXL 2024-10-18 20:57:13 +09:00
kohya-ss
c80c304779 Refactor caching in train scripts 2024-10-12 20:18:41 +09:00