Commit message (Expand) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | UpdateHEADmaster | Volpeon | 2023-06-25 | 1 | -1/+0 |
* | Update | Volpeon | 2023-06-25 | 1 | -3/+0 |
* | New loss scaling | Volpeon | 2023-06-24 | 1 | -67/+41 |
* | Added Prodigy optimizer | Volpeon | 2023-06-24 | 1 | -1/+9 |
* | Fix | Volpeon | 2023-06-24 | 1 | -4/+5 |
* | Fixes | Volpeon | 2023-06-24 | 1 | -2/+5 |
* | Remove training guidance_scale | Volpeon | 2023-06-22 | 1 | -17/+1 |
* | Remove convnext | Volpeon | 2023-06-22 | 1 | -13/+0 |
* | Update | Volpeon | 2023-06-21 | 1 | -13/+0 |
* | Update | Volpeon | 2023-06-21 | 1 | -72/+149 |
* | Fix LoRA training with DAdan | Volpeon | 2023-05-16 | 1 | -2/+2 |
* | Avoid model recompilation due to varying prompt lengths | Volpeon | 2023-05-16 | 1 | -2/+2 |
* | Patch xformers to cast dtypes | Volpeon | 2023-05-16 | 1 | -6/+5 |
* | Support LoRA training for token embeddings | Volpeon | 2023-05-16 | 1 | -9/+3 |
* | Update | Volpeon | 2023-05-11 | 1 | -9/+15 |
* | Update | Volpeon | 2023-05-06 | 1 | -8/+12 |
* | Update | Volpeon | 2023-05-05 | 1 | -12/+21 |
* | torch.compile won't work yet, keep code prepared | Volpeon | 2023-04-29 | 1 | -0/+3 |
* | Fixed model compilation | Volpeon | 2023-04-29 | 1 | -3/+3 |
* | Support torch.compile | Volpeon | 2023-04-28 | 1 | -4/+8 |
* | Fixed loss/acc logging | Volpeon | 2023-04-28 | 1 | -16/+16 |
* | Update | Volpeon | 2023-04-27 | 1 | -26/+32 |
* | Update | Volpeon | 2023-04-21 | 1 | -2/+2 |
* | Fix | Volpeon | 2023-04-16 | 1 | -1/+1 |
* | Improved automation caps | Volpeon | 2023-04-16 | 1 | -8/+9 |
* | Fix | Volpeon | 2023-04-16 | 1 | -4/+3 |
* | TI via LoRA | Volpeon | 2023-04-15 | 1 | -2/+9 |
* | Added cycle LR decay | Volpeon | 2023-04-13 | 1 | -1/+4 |
* | Update | Volpeon | 2023-04-13 | 1 | -18/+17 |
* | Experimental convnext discriminator support | Volpeon | 2023-04-11 | 1 | -24/+59 |
* | Store sample images in Tensorboard as well | Volpeon | 2023-04-11 | 1 | -13/+12 |
* | Update | Volpeon | 2023-04-11 | 1 | -12/+10 |
* | Update | Volpeon | 2023-04-10 | 1 | -24/+27 |
* | Fix sample gen: models sometimes weren't in eval mode | Volpeon | 2023-04-10 | 1 | -0/+3 |
* | Update | Volpeon | 2023-04-10 | 1 | -3/+1 |
* | Update | Volpeon | 2023-04-09 | 1 | -2/+2 |
* | Fix | Volpeon | 2023-04-09 | 1 | -1/+1 |
* | Update | Volpeon | 2023-04-08 | 1 | -35/+22 |
* | Fix | Volpeon | 2023-04-07 | 1 | -6/+0 |
* | Update | Volpeon | 2023-04-06 | 1 | -36/+4 |
* | MinSNR code from diffusers | Volpeon | 2023-04-05 | 1 | -19/+30 |
* | Improved slerp noise offset: Dedicated black image instead of negative offset | Volpeon | 2023-04-05 | 1 | -72/+40 |
* | New offset noise test | Volpeon | 2023-04-05 | 1 | -8/+89 |
* | TI: Bring back old embedding decay | Volpeon | 2023-04-04 | 1 | -2/+2 |
* | TI: Delta learning | Volpeon | 2023-04-03 | 1 | -2/+2 |
* | Update dataset format: Separate prompt and keywords | Volpeon | 2023-04-02 | 1 | -3/+6 |
* | Update | Volpeon | 2023-04-02 | 1 | -3/+3 |
* | Revert | Volpeon | 2023-04-01 | 1 | -1/+1 |
* | Update | Volpeon | 2023-04-01 | 1 | -5/+13 |
* | Add support for Adafactor, add TI initializer noise | Volpeon | 2023-04-01 | 1 | -1/+2 |