search

ogkalu2 / Merge-Stable-Diffusion-models-without-distortion

Adaptation of the merging method described in the paper - Git Re-Basin: Merging Models modulo Permutation Symmetries (https://arxiv.org/abs/2209.04836) for Stable Diffusion
MIT License
139 stars 21 forks source link

KeyError: 'model_ema.diffusion_modeltime_embed0weight' #19

Open NaughtDZ opened 1 year ago

NaughtDZ commented 1 year ago

Using half precision

---------------------
     ITERATION 1
---------------------

new alpha = 0.05

Traceback (most recent call last): File "I:\Dreambooth-Stable-Diffusion\Merge-Stable-Diffusion-models-without-distortion\SD_rebasin_merge.py", line 55, in theta_0 = {key: (1 - (new_alpha)) theta_0[key] + (new_alpha) value for key, value in theta_1.items() if "model" in key and key in theta_1} File "I:\Dreambooth-Stable-Diffusion\Merge-Stable-Diffusion-models-without-distortion\SD_rebasin_merge.py", line 55, in theta_0 = {key: (1 - (new_alpha)) theta_0[key] + (new_alpha) value for key, value in theta_1.items() if "model" in key and key in theta_1} KeyError: 'model_ema.diffusion_modeltime_embed0weight'

vaguenebula commented 1 year ago

what models are you trying to merge?

NaughtDZ commented 1 year ago

what models are you trying to merge?

Anything-V3.0 and wd-v1-3-float16

wilhelmw201 commented 1 year ago

maybe you can try removing the ema first with a tool like this one https://github.com/Akegarasu/sd-model-converter