aboutsummaryrefslogtreecommitdiffstats
path: root/modules/sd_samplers_kdiffusion.py
AgeCommit message (Collapse)AuthorLines
2023-08-12put refiner into main UI, into the new accordions sectionAUTOMATIC1111-1/+2
add VAE from main model into infotext, not from refiner model option to make scripts UI without gr.Group fix inconsistencies with refiner when usings samplers that do more denoising than steps
2023-08-10lintAUTOMATIC1111-1/+1
2023-08-10return the line lost during the mergeAUTOMATIC1111-0/+1
2023-08-10resolve merge issuesAUTOMATIC1111-2/+1
2023-08-10Merge branch 'dev' into refinerAUTOMATIC1111-1/+2
2023-08-08by request: fix tiled vae extensionAUTOMATIC1111-2/+3
2023-08-08merge errorsAUTOMATIC1111-5/+12
2023-08-08Merge branch 'dev' into refinerAUTOMATIC1111-358/+21
2023-08-08put commonly used samplers on top, make DPM++ 2M Karras the default choiceAUTOMATIC1111-4/+4
2023-08-08REMOVEAUTOMATIC1111-2/+1
2023-08-08add CFG denoiser implementation for DDIM, PLMS and UniPC (this is the commit ↵AUTOMATIC1111-138/+14
when you can run both old and new implementations to compare them)
2023-08-08split sd_samplers_kdiffusion into twoAUTOMATIC1111-189/+2
2023-08-08Split history: mv temp modules/sd_samplers_kdiffusion.pyAUTOMATIC1111-0/+511
2023-08-08Split history: mv modules/sd_samplers_kdiffusion.py tempAUTOMATIC1111-511/+0
2023-08-06add infotextAUTOMATIC1111-8/+16
add proper support for recalculating conds in k-diffusion samplers remove support for compvis samplers
2023-08-06lint!AUTOMATIC1111-1/+1
2023-08-06initial refiner supportAUTOMATIC1111-6/+24
2023-08-05Assume 0 = inf for s_tmaxcatboxanon-1/+1
2023-08-05Read kdiffusion sigma params from optscatboxanon-7/+7
2023-08-05Add sigma params to infotextcatboxanon-1/+33
2023-08-04Add exponential scheduler for DPM-Solver++(2M) SDEcatboxanon-0/+4
Better quality results than Karras. Related discussion: https://gist.github.com/crowsonkb/3ed16fba35c73ece7cf4b9a2095f2b78
2023-08-03add NV option for Random number generator source setting, which allows to ↵AUTOMATIC1111-4/+1
generate same pictures on CPU/AMD/Mac as on NVidia videocards.
2023-07-29split the new sampler into a different fileAUTOMATIC1111-71/+4
2023-07-29Split history: mv temp modules/sd_samplers_kdiffusion.pyAUTOMATIC1111-0/+545
2023-07-29Split history: mv modules/sd_samplers_kdiffusion.py tempAUTOMATIC1111-545/+0
2023-07-29Merge pull request #11850 from lambertae/restart_samplingAUTOMATIC1111-2/+71
Restart sampling
2023-07-25fix progress bar & torchHijacklambertae-6/+13
2023-07-20new restart schemelambertae-1/+4
2023-07-20fix rufflambertae-1/+1
2023-07-20 allow choise of restart_list & use karras from kdiffusionlambertae-20/+12
2023-07-20restart-sampler with correct stepslambertae-8/+14
2023-07-18code stylinglambertae-9/+5
2023-07-18fix floating errorlambertae-5/+6
2023-07-18remove useless headerlambertae-2/+0
2023-07-18add restart samplerlambertae-2/+68
2023-07-12SDXL supportAUTOMATIC1111-1/+1
2023-07-11getting SD2.1 to run on SDXL repoAUTOMATIC1111-10/+35
2023-06-27add missing infotext entry for the pad cond/uncond optionAUTOMATIC-1/+10
2023-05-31Merge pull request #10638 from catboxanon/patch/revert-10586AUTOMATIC1111-2/+2
Revert discarding penultimate sigma for DPM-Solver++(2M) SDE
2023-05-27updates for the noise schedule settingsAUTOMATIC-14/+16
2023-05-24use Schedule instead of SchedKohaku-Blueleaf-4/+4
2023-05-24improvementsKohaku-Blueleaf-10/+17
See: https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/10649#issuecomment-1561047723
2023-05-24Use automatic instead of None/defaultKohaku-Blueleaf-3/+3
2023-05-23Use type to determine if it is enableKohaku-Blueleaf-3/+3
2023-05-23Use better way to implKohaku-Blueleaf-6/+13
2023-05-23Better hint for userKohaku-Blueleaf-2/+3
Co-authored-by: catboxanon <122327233+catboxanon@users.noreply.github.com>
2023-05-22Revert #10586catboxanon-2/+2
2023-05-23use sigma_max/min in model if sigma_max/min is 0Kohaku-Blueleaf-2/+3
2023-05-23Add error information for recursion errorKohaku-Blueleaf-0/+6
2023-05-22remove not related codeKohaku-Blueleaf-1/+0