summaryrefslogtreecommitdiffstats
path: root/environment.yaml
Commit message (Expand)AuthorAgeFilesLines
* Added Prodigy optimizerVolpeon2023-06-241-0/+1
* UpdateVolpeon2023-06-211-5/+6
* Fix LoRA training with DAdanVolpeon2023-05-161-1/+1
* Update depsVolpeon2023-05-161-3/+3
* UpdateVolpeon2023-05-111-3/+5
* torch.compile won't work yet, keep code preparedVolpeon2023-04-291-2/+1
* Support torch.compileVolpeon2023-04-281-5/+9
* UpdateVolpeon2023-04-111-1/+1
* Add color jitterVolpeon2023-04-051-0/+1
* Bring back Lion optimizerVolpeon2023-04-031-0/+1
* FixVolpeon2023-03-311-3/+3
* FixVolpeon2023-03-221-1/+1
* Added dadaptationVolpeon2023-03-211-1/+1
* Fixed SNR weighting, re-enabled xformersVolpeon2023-03-211-0/+2
* Test: https://arxiv.org/pdf/2303.09556.pdfVolpeon2023-03-171-24/+22
* Implemented different noise offsetVolpeon2023-03-031-1/+1
* UpdateVolpeon2023-03-011-1/+1
* Embedding normalization: Ignore tensors with grad = 0Volpeon2023-02-211-2/+1
* Added Lion optimizerVolpeon2023-02-171-0/+1
* Back to xformersVolpeon2023-02-171-3/+6
* Remove xformers, switch to Pytorch NightlyVolpeon2023-02-171-11/+7
* Add LoraVolpeon2023-02-071-2/+2
* Restored LR finderVolpeon2023-01-201-1/+1
* Fixed Conda envVolpeon2023-01-151-2/+4
* TI: Use grad clipping from LoRA #104Volpeon2023-01-111-1/+1
* UpdateVolpeon2023-01-061-2/+2
* Package updateVolpeon2023-01-061-1/+1
* Added multi-vector embeddingsVolpeon2022-12-311-0/+1
* Added learning rate finderVolpeon2022-12-271-0/+1
* Improved class prompt handlingVolpeon2022-12-231-1/+1
* Dependency cleanup/upgradesVolpeon2022-12-201-24/+10
* Package updatesVolpeon2022-12-111-3/+3
* UpdateVolpeon2022-12-011-1/+1
* UpdateVolpeon2022-11-271-2/+2
* UpdateVolpeon2022-11-031-2/+3
* Various updatesVolpeon2022-10-121-2/+2
* UpdateVolpeon2022-10-081-1/+1
* Incorporate upstream changesVolpeon2022-09-271-0/+1
* Use diffusers fork with Flash AttentionVolpeon2022-09-271-1/+3
* Better sample filenames, optimizationsVolpeon2022-09-271-6/+4
* InitVolpeon2022-09-261-0/+36