1580158
/
kohya_ss
mirror of https://github.com/bmaltais/kohya_ss
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
2,902 Commits
16 Branches
178 Tags
68 MiB
Commit Graph

160 Commits (d8a51f34fd15dc4fbbe4e8416cd3270f23818c09)

Author SHA1 Message Date
bmaltais d8a51f34fd Set `max_train_steps` to 0 if not specified in older `.json` config files 2024-04-26 07:11:07 -04:00
bmaltais 1b71c7f5ee Fix issue where tensorboard was displayed when tensorflow was not installed 2024-04-25 18:39:06 -04:00
bmaltais 433fabf7d8 Fix [24.0.6] Train toml config seed type error #2370 2024-04-25 13:10:55 -04:00
bmaltais 1f5f3cf027 Prevent tkinter import crash 2024-04-25 11:20:27 -04:00
bmaltais 81cd49830f Convert gradio_resize_lora_tab to shell false 2024-04-21 08:42:52 -04:00
bmaltais 214f7199a3 Fix issue with verify lora tool 2024-04-21 08:29:00 -04:00
bmaltais 8234e52ded
Add validation of lr scheduler and optimizer arguments (#2358) 2024-04-20 15:53:11 -04:00
bmaltais 58e57a3365
Fix issue with lora_network_weights not being loaded (#2357) 2024-04-20 15:05:35 -04:00
bmaltais 4923f5647c
Make Start/Stop buttons visible in headless (#2356) 2024-04-20 14:27:19 -04:00
bmaltais 5a801649c2 Convert config with use_wandb to log_with = wandb 2024-04-19 13:16:23 -04:00
Maatra 25d7c6889f
Changed logger checkbox to dropdown, renamed use_wandb -> log_with (#2352) 2024-04-19 13:14:15 -04:00
bmaltais 542af98980 Fix wrong default for max_train_steps and max_train_epoch 2024-04-19 13:04:02 -04:00
bmaltais 4c5f4dfd27 update file read and write to utf-8 2024-04-19 10:50:43 -04:00
bmaltais 680ea75c0b
2346 wd14 captioning issue with undesired tags nor tag replacement (#2350) 
* Improve how the code handle v3 (and other v2) models download and use

* Switch to shell=False always
 2024-04-19 10:31:06 -04:00
bmaltais 40c5ac6e52 Hide tensorboard button if tensorflow module is not installed 2024-04-19 08:11:30 -04:00
bmaltais 4df07c4a73 Write toml as UTF-8 2024-04-18 22:34:06 -04:00
bmaltais 98d826c5ec
Fix bad Lora parameters (#2341) 2024-04-18 22:21:22 -04:00
bmaltais b4f1589f18
2330 another seemingly new data validation leads to unusable configs 2403 (#2339) 
* Fix issue with None values staying as 0 in toml config file

* Fix noise_offset issues (and possibly other issues)
 2024-04-18 21:47:00 -04:00
bmaltais 07db163be9
2325 quotes wrapping python path cause subprocess cant find target in v2403 (#2338) 
* Fix issue with doublequotes wrapped python path

* Fix path. Always assume no space in path to kohya_ss repo

* Add img path with space

* Enforce no white space in path to kohya_ss
 2024-04-18 20:45:43 -04:00
b-fission 415c8fb5c1 Use correct file extensions when browsing for model file 2024-04-17 22:17:02 -05:00
bmaltais 85bdb37789 Improve PYTHON variable 2024-04-17 20:09:06 -04:00
bmaltais 84fb7ce8c9 Update utility paths and print toml file when clicking the Print training command button 2024-04-17 19:54:56 -04:00
bmaltais 962fc681cc Fix issue with sample prompt creation 2024-04-17 17:45:59 -04:00
Bernard Maltais be5d30130f Fix clip_skip issue 2024-04-17 13:25:55 -04:00
bmaltais 1898851aa6 Fix bad merge 2024-04-17 07:32:41 -04:00
bmaltais aa4d13fc93 Merge branch 'master' of https://github.com/bmaltais/kohya_ss into dev 2024-04-17 07:20:50 -04:00
bmaltais 43087f7b3d Update training info display 2024-04-17 07:11:40 -04:00
bmaltais 8bda4f203c Align toml file content to sd-scripts defaults 2024-04-16 21:17:53 -04:00
bmaltais 13b836ad76 Fix issues with steps calculations 2024-04-16 14:52:20 -04:00
bmaltais dc5d1c9018 Add debiased estimation loss to finetune 2024-04-16 13:27:02 -04:00
bmaltais 79cbde3329 Fix localization 2024-04-16 09:00:59 -04:00
bmaltais 9037fd0373
Improve process execution under Linux (and windows) (#2301) 
* Update how process is executed

* Auto config use_shell=True on posix systems
 2024-04-16 07:53:44 -04:00
bmaltais e6a8dec98d
Implement use_shell as parameter (#2297) 2024-04-15 20:26:09 -04:00
bmaltais f4658f9e01 Fix tkinter issue for headless 2024-04-15 14:07:08 -04:00
bmaltais e519ffec06 Fix warning about dropdown 2024-04-15 13:59:13 -04:00
bmaltais a22d4622cd
Add support for metadata parameters (#2295) 2024-04-15 13:26:39 -04:00
bmaltais a8320e3f2c Add support for accelerate dynamo parameters 2024-04-15 10:31:05 -04:00
Bernard Maltais cd1c46e10f Fix issue with extra_accelerate_launch_args 2024-04-15 09:24:10 -04:00
bmaltais 008ec2b63a Fix extra_accelerate_launch_args args parsing issue 2024-04-15 09:22:48 -04:00
bmaltais b9eefb2370 Cleanup old commented code 2024-04-15 09:16:47 -04:00
bmaltais 098be177e3 Fix wd14 caption accelerate path 2024-04-15 08:53:30 -04:00
bmaltais c4e0655a7a Fix accelerate path for linux systems 2024-04-15 08:50:48 -04:00
bmaltais b64d856340
Implement LR Warmup disabling when constant scheduler is used (#2292) 
* Implement LR Warmup disabling when constant scheduler is used
 2024-04-15 08:37:16 -04:00
bmaltais 18ef5305d2 Make gradient_accumulation_steps an int 2024-04-15 08:10:24 -04:00
bmaltais eec6f9baf4 Convert str numbers to proper int or float 2024-04-15 08:02:39 -04:00
bmaltais 5316db3fa7 Add huggingface parms to config file 2024-04-14 22:03:36 -04:00
bmaltais 91b679f9ca Update dreambooth attributes 2024-04-14 21:41:04 -04:00
bmaltais a08c7c6ef1 Update parameters handling 2024-04-14 19:10:26 -04:00
bmaltais e98fc40a63
- Implement HuggingFace inputs in all training tabs (#2287) 
- Switch to sd-scripts configuration files to pass most of the parameters in a secure way.
 2024-04-14 18:43:45 -04:00
bmaltais fa5c65eade Update readme 2024-04-13 21:22:52 -04:00