summaryrefslogtreecommitdiffstats
path: root/train_dreambooth.py
Commit message (Collapse)AuthorAgeFilesLines
...
* More modularizationVolpeon2023-01-131-207/+65
|
* Removed PromptProcessor, modularized training loopVolpeon2023-01-131-5/+2
|
* Added TI decay start offsetVolpeon2023-01-131-1/+1
|
* Code deduplicationVolpeon2023-01-131-56/+15
|
* UpdateVolpeon2023-01-121-14/+42
|
* TI: Use grad clipping from LoRA #104Volpeon2023-01-111-2/+2
|
* Better defaultsVolpeon2023-01-101-3/+2
|
* FixVolpeon2023-01-101-1/+1
|
* Added arg to disable tag shufflingVolpeon2023-01-101-1/+8
|
* Enable buckets for validation, fixed vaildation repeat argVolpeon2023-01-091-4/+1
|
* Add --valid_set_repeatVolpeon2023-01-091-0/+10
|
* Improved aspect ratio bucketingVolpeon2023-01-081-0/+27
|
* Fixed aspect ratio bucketing; allow passing token IDs to pipelineVolpeon2023-01-081-6/+8
|
* Improved aspect ratio bucketingVolpeon2023-01-081-55/+45
|
* CleanupVolpeon2023-01-071-134/+131
|
* Added progressive aspect ratio bucketingVolpeon2023-01-071-9/+3
|
* UpdateVolpeon2023-01-071-3/+3
|
* Added EMA to TIVolpeon2023-01-051-1/+1
|
* UpdateVolpeon2023-01-051-7/+7
|
* FixVolpeon2023-01-051-2/+2
|
* Various cleanupsVolpeon2023-01-051-65/+21
|
* Various updatesVolpeon2023-01-041-0/+8
|
* Better eval generatorVolpeon2023-01-041-4/+4
|
* Fixed reproducibility, more consistant validationVolpeon2023-01-041-5/+19
|
* Don't use vector_dropout by defaultVolpeon2023-01-031-1/+1
|
* Added vector dropoutVolpeon2023-01-031-5/+19
|
* Fixed LR finderVolpeon2023-01-021-2/+2
|
* UpdateVolpeon2023-01-021-4/+12
|
* FixVolpeon2023-01-021-4/+0
|
* Save args before training, tooVolpeon2023-01-021-0/+4
|
* FixVolpeon2023-01-021-2/+1
|
* UpdateVolpeon2023-01-021-7/+6
|
* CleanupVolpeon2023-01-011-3/+4
|
* UpdatesVolpeon2023-01-011-104/+124
|
* Fixed accuracy calc, other improvementsVolpeon2023-01-011-1/+29
|
* Misc improvementsVolpeon2022-12-301-29/+18
|
* Training script improvementsVolpeon2022-12-301-3/+38
|
* Integrated updates from diffusersVolpeon2022-12-281-10/+6
|
* Improved learning rate finderVolpeon2022-12-271-3/+2
|
* Added learning rate finderVolpeon2022-12-271-75/+54
|
* Set default dimensions to 768; add config inheritanceVolpeon2022-12-261-4/+3
|
* Code simplifications, avoid autocastVolpeon2022-12-251-0/+12
|
* UpdateVolpeon2022-12-251-1/+0
|
* Fixed Textual InversionVolpeon2022-12-241-1/+8
|
* Training updateVolpeon2022-12-241-16/+4
|
* UpdateVolpeon2022-12-241-1/+9
|
* Better dataset prompt handlingVolpeon2022-12-231-1/+1
|
* num_class_images is now class images per train imageVolpeon2022-12-231-1/+1
|
* Improved class prompt handlingVolpeon2022-12-231-22/+8
|
* Moved common training code into separate moduleVolpeon2022-12-211-110/+16
|