summaryrefslogtreecommitdiffstats
path: root/infer.py
Commit message (Expand)AuthorAgeFilesLines
* Pipeline: Perlin noise for init imageVolpeon2023-03-041-1/+1
* UpdateVolpeon2023-02-181-0/+2
* Added Lion optimizerVolpeon2023-02-171-2/+5
* Inference script: Better scheduler configVolpeon2023-02-171-19/+37
* Back to xformersVolpeon2023-02-171-2/+2
* Remove xformers, switch to Pytorch NightlyVolpeon2023-02-171-3/+4
* Integrated WIP UniPC schedulerVolpeon2023-02-161-3/+12
* UpdateVolpeon2023-02-131-7/+7
* Add LoraVolpeon2023-02-071-3/+18
* WIP: Modularization ("free(): invalid pointer" my ass)Volpeon2023-01-141-4/+15
* CleanupVolpeon2023-01-141-15/+4
* Added progressive aspect ratio bucketingVolpeon2023-01-071-11/+12
* Add prompt template argument to inferenceVolpeon2023-01-061-0/+7
* UpdateVolpeon2023-01-051-1/+1
* UpdateVolpeon2023-01-051-11/+29
* Various cleanupsVolpeon2023-01-051-3/+3
* UpdateVolpeon2023-01-041-4/+3
* Various updatesVolpeon2023-01-041-12/+44
* Bugfixes for multi-vector token handlingVolpeon2022-12-311-4/+9
* Added multi-vector embeddingsVolpeon2022-12-311-5/+9
* Set default dimensions to 768; add config inheritanceVolpeon2022-12-261-6/+5
* Code simplifications, avoid autocastVolpeon2022-12-251-24/+24
* Avoid increased VRAM usage on validationVolpeon2022-12-151-2/+2
* More generic datset filterVolpeon2022-12-141-2/+3
* Unified loading of TI embeddingsVolpeon2022-12-131-32/+2
* Add support for resume in Textual InversionVolpeon2022-12-131-1/+1
* Dreambooth: Support loading Textual Inversion embeddingsVolpeon2022-12-121-1/+1
* Various updated; shuffle prompt content during trainingVolpeon2022-12-101-5/+22
* Fix sample stepsVolpeon2022-12-061-1/+1
* UpdateVolpeon2022-12-011-0/+1
* UpdateVolpeon2022-12-011-1/+2
* UpdateVolpeon2022-11-271-29/+23
* RefactoringVolpeon2022-11-141-3/+4
* UpdateVolpeon2022-11-141-8/+13
* UpdateVolpeon2022-11-031-1/+1
* New Euler_a schedulerVolpeon2022-10-261-12/+4
* Add optional TI functionality to DreamboothVolpeon2022-10-211-3/+3
* Improved prompt handlingVolpeon2022-10-181-10/+5
* Implemented extended prompt limitVolpeon2022-10-171-1/+11
* Removed aesthetic gradients; training improvementsVolpeon2022-10-141-28/+4
* Added support for Aesthetic GradientsVolpeon2022-10-131-23/+52
* Added TI+Dreambooth trainingVolpeon2022-10-131-4/+4
* Various updatesVolpeon2022-10-121-5/+5
* UpdateVolpeon2022-10-091-2/+2
* UpdateVolpeon2022-10-081-5/+7
* Training: Create multiple class images per training imageVolpeon2022-10-071-1/+4
* UpdateVolpeon2022-10-061-0/+14
* Inference: Add support for embeddingsVolpeon2022-10-041-2/+30
* Small perf improvementsVolpeon2022-10-031-2/+3
* Small performance improvementsVolpeon2022-10-021-5/+13