bmaltais
ba2b3ab0db
Merge pull request #711 from ertugrul-dmr/master
...
Fix for UnicodeDecodeError
2023-05-04 14:18:00 -04:00
ertugrul-dmr
4740ad0351
fix for UnicodeDecodeError
2023-05-04 16:43:32 +03:00
bmaltais
d3a89443cd
Add multi-resolution noise training support
2023-05-03 09:59:18 -04:00
bmaltais
111527b490
Add support for Lion8bit
2023-05-03 08:49:48 -04:00
bmaltais
be3ba65127
Add missing options to WD14 tagger
...
Refactor code
2023-05-02 09:50:18 -04:00
bmaltais
856d610ea9
Add --cache_latents_to_disk option to the GUI
2023-05-01 20:10:47 -04:00
bmaltais
7bd1cb9d08
Merge branch 'main' of https://github.com/kohya-ss/sd-scripts into dev
2023-04-27 09:03:59 -04:00
bmaltais
d9d4789c6a
fix wront type for caption_dropout_rate
2023-04-07 09:03:59 -04:00
bmaltais
1f153629fe
Fix bug 554
2023-04-07 08:55:03 -04:00
bmaltais
7b87edfd29
Adding GUI support for new sd-scripts features
2023-04-05 08:33:24 -04:00
bmaltais
54a204165b
Update to latest kohya_ss code
2023-04-04 18:05:20 -04:00
bmaltais
369341d69f
Revert "macOS GUI functionality, sub-processed GUI components"
2023-04-01 21:28:47 -04:00
JSTayco
eef5becab8
All sorts of broken, but I need to commit this for now so I don't lose it. WIP: Using some OOP to reduce imports and centralize some code. No need to remake Tk Windows everywhere for example. Renamed common_gui to common_gui_functions.py to make some of the new code separation more obvious.
2023-03-31 14:39:10 -07:00
JSTayco
e5b83df675
Removed one warning dealing with get_file_path()
...
Using lambdas now to pass in variable amount of arguments from components. This works right now with a few open windows, but saving and possibly loading will be broken right now. They need the lambda treatment next.
I also split the JSON validation placeholder to library/common_utilities.py.
2023-03-30 13:13:25 -07:00
jstayco
160e371be3
Merge branch 'dev' into macos_gui
2023-03-30 01:53:32 -07:00
JSTayco
7b5639cff5
Huge WIP
...
This is a massive WIP and should not be trusted or used right now. However, major milestones have been crossed. Both message boxes and file dialogs are now properly subprocessed and work on macOS. I think by extension, it may work on runpod environments as well, but that remains to be tested.
2023-03-30 01:40:00 -07:00
bmaltais
13c4b1f73b
v21.3.7
2023-03-29 19:43:23 -04:00
bmaltais
8822eab5a6
Merge new sd-scripts updates
2023-03-28 11:54:42 -04:00
bmaltais
b6332ceb0f
Fix missing macos requirements file
2023-03-26 06:47:26 -04:00
bmaltais
c38c018197
Prevent the file selection GUI from running
2023-03-25 20:29:04 -04:00
bmaltais
7e7a8b6aab
Fix issue with msgbox
2023-03-25 12:39:02 -04:00
bmaltais
d0ffba6650
Remove bad save_model_as choices for LoRA and TI
2023-03-25 09:08:02 -04:00
bmaltais
a9aafff259
Fix issue 433
2023-03-25 07:03:31 -04:00
bmaltais
f83638b8a6
Add device support
2023-03-22 12:55:30 -04:00
bmaltais
1c8d901c3b
Update to latest sd-scripts updates
2023-03-21 20:20:57 -04:00
bmaltais
ccae80186a
Update to latest sd-script code
2023-03-20 08:47:00 -04:00
bmaltais
baf009d2b1
Fix basic captioning logic
2023-03-15 19:31:52 -04:00
bmaltais
5597f19cbc
Merge branch 'dev' into LoHa-Support
2023-03-11 10:47:40 -05:00
bmaltais
a65555ea67
Add support to load a config without opening the UI to get the file name
2023-03-10 20:05:38 -05:00
bmaltais
ba9e26a99d
LoHa support
2023-03-09 07:49:50 -05:00
bmaltais
2eb7b3bdc3
Fix issues with code detected while testing
2023-03-08 21:16:54 -05:00
bmaltais
532552650b
Fix issue 335
2023-03-08 19:33:53 -05:00
bmaltais
7ed8f7c3c5
Add additional parameters feild
2023-03-07 07:42:13 -05:00
bmaltais
9e6b4cb69b
Update presets
2023-03-05 21:42:28 -05:00
bmaltais
dbf959db68
Add logic to v2 checkbox
2023-03-05 21:38:20 -05:00
bmaltais
cc7aee2301
Improve custom preset handling
2023-03-05 21:10:24 -05:00
bmaltais
09939ff8a8
Remove legacy 8bit adam checkbox
2023-03-05 10:34:09 -05:00
bmaltais
3beeef4414
Add linux support
2023-03-04 18:56:22 -05:00
bmaltais
e93de7f8ea
Fix issue 296
2023-03-04 17:46:32 -05:00
bmaltais
4c1448be72
Fix again the custom model config load
...
Update resize lora
2023-03-03 22:08:06 -05:00
bmaltais
d30abe5491
Fix issue 277
2023-03-03 07:11:15 -05:00
bmaltais
4c762e3abb
Upgrade Gradio
...
Fix Gradio issues
Fix missing LoCon new GUI
2023-03-02 20:39:07 -05:00
bmaltais
182080bb78
Run blue
2023-03-01 19:24:11 -05:00
bmaltais
7f0e5683c6
v21.0.1
2023-03-01 19:02:04 -05:00
bmaltais
1e3055c895
Update tensorboard
2023-03-01 13:14:47 -05:00
bmaltais
9d2e3f85a2
Add tensorboard support
2023-02-26 19:49:22 -05:00
bmaltais
6b5d6303cc
Update
2023-02-26 15:11:21 -05:00
bmaltais
f213b15014
Updates
2023-02-24 20:37:51 -05:00
bmaltais
8775667fc7
Updates
2023-02-23 21:48:45 -05:00
bmaltais
bf0344ba9e
Adding GUI support for new features
2023-02-22 20:32:57 -05:00
bmaltais
758bfe85dc
Adding support for Lion optimizer in gui
2023-02-19 20:13:03 -05:00
bmaltais
e5f8ba559f
Add server_port and inbrowser support
...
- to all gui scripts
2023-02-10 08:22:03 -05:00
bmaltais
09d3a72cd8
Adding support for caption dropout
2023-02-07 20:58:35 -05:00
bmaltais
8d559ded18
* 2023/02/06 (v20.7.0)
...
- ``--bucket_reso_steps`` and ``--bucket_no_upscale`` options are added to training scripts (fine tuning, DreamBooth, LoRA and Textual Inversion) and ``prepare_buckets_latents.py``.
- ``--bucket_reso_steps`` takes the steps for buckets in aspect ratio bucketing. Default is 64, same as before.
- Any value greater than or equal to 1 can be specified; 64 is highly recommended and a value divisible by 8 is recommended.
- If less than 64 is specified, padding will occur within U-Net. The result is unknown.
- If you specify a value that is not divisible by 8, it will be truncated to divisible by 8 inside VAE, because the size of the latent is 1/8 of the image size.
- If ``--bucket_no_upscale`` option is specified, images smaller than the bucket size will be processed without upscaling.
- Internally, a bucket smaller than the image size is created (for example, if the image is 300x300 and ``bucket_reso_steps=64``, the bucket is 256x256). The image will be trimmed.
- Implementation of [#130 ](https://github.com/kohya-ss/sd-scripts/issues/130 ).
- Images with an area larger than the maximum size specified by ``--resolution`` are downsampled to the max bucket size.
- Now the number of data in each batch is limited to the number of actual images (not duplicated). Because a certain bucket may contain smaller number of actual images, so the batch may contain same (duplicated) images.
- ``--random_crop`` now also works with buckets enabled.
- Instead of always cropping the center of the image, the image is shifted left, right, up, and down to be used as the training data. This is expected to train to the edges of the image.
- Implementation of discussion [#34 ](https://github.com/kohya-ss/sd-scripts/discussions/34 ).
2023-02-06 11:04:07 -05:00
bmaltais
cbfc311687
Integrate new bucket parameters in GUI
2023-02-05 20:07:00 -05:00
bmaltais
2626214f8a
Add support for LoRA resizing
2023-02-04 11:55:06 -05:00
bmaltais
c8f4c9d6e8
Add support for lr_scheduler_num_cycles, lr_scheduler_power
2023-01-30 08:26:15 -05:00
bmaltais
202923b3ce
Add support for --keep_token option
2023-01-27 07:33:44 -05:00
bmaltais
03bd2e9b01
Add TI training support
2023-01-26 16:22:58 -05:00
bmaltais
49bada0d25
Update default save model file format to safetensors
2023-01-22 18:21:09 -05:00
bmaltais
fcad6bfd98
Add default for caption extension tools
2023-01-20 18:41:37 -05:00
bmaltais
567221549f
Complete training code refactoring
2023-01-15 19:59:40 -05:00
bmaltais
123cf4e3c5
Move cxommon adv train parm to common gui
2023-01-15 15:03:04 -05:00
bmaltais
abccecb093
Move save_state and resume to common gui
...
Format code
2023-01-15 12:01:59 -05:00
bmaltais
6aed2bb402
Add support for new arguments:
...
- max_train_epochs
- max_data_loader_n_workers
Move some of the codeto common gui library.
2023-01-15 11:05:22 -05:00
bmaltais
dc5afbb057
Move functions to common_gui
...
Add model name support
2023-01-09 11:48:57 -05:00
bmaltais
c79a8a9409
Add find/replace option to Basic Caption utility
2023-01-08 19:12:01 -05:00
bmaltais
34f7cd8e57
Add new Utility to Extract a LoRA from a finetuned model
2023-01-06 18:25:55 -05:00
bmaltais
1d460a09fd
add support for color and flip augmentation to "Dreambooth LoRA"
2023-01-01 22:43:44 -05:00
bmaltais
b44f075f60
Implement open and save config for LoRA
2022-12-29 14:00:02 -05:00
bmaltais
f60c8addd5
Making gui more modular by dividing it
2022-12-22 11:51:34 -05:00
bmaltais
706dfe157f
Merge dreambooth and finetuning in one repo to align with kohya_ss new repo ( #10 )
...
* Merge both dreambooth and finetune back in one repo
2022-12-20 09:15:17 -05:00
bmaltais
1f1dd5c4de
Add support for:
...
shuffle_caption,
save_state,
resume,
prior_loss_weight,
Fix issue with config open and save
2022-12-19 21:50:05 -05:00
bmaltais
1d412726b3
Add new prefix and postfic for captions
2022-12-19 11:43:29 -05:00
bmaltais
61a9481903
Fix save as issue
2022-12-19 09:47:35 -05:00
bmaltais
c90aa2cc61
- Fix file/folder opening behind the browser window
...
- Add WD14 and BLIP captioning to utilities
- Improve overall GUI layout
2022-12-19 09:22:52 -05:00
bmaltais
0ca93a7aa7
v18.1: Model conversion utility
2022-12-18 13:11:10 -05:00