I'm sorry to use an issue for this. My C: drive is very full and the hugginface cache is using about 30Gb so I want to move it to one of my bigger drives. On SD webui I managed to do it by doing a set for the HF_HOME, I thought that this...
File “/content/kohya_ss/library/config_util.py”, line 494, in generate_dataset_group_by_blueprint dataset.make_buckets() File “/content/kohya_ss/library/train_util.py”, line 738, in make_buckets info.image_size = self.get_image_size(info.absolute_path) File “/content/kohya_ss/...
As LoRA becomes more and more popular recently, while the communities get used to share LoRA file in safetensors or ckpt format, I think it's quite important that we have an convenient approach/script to convert the LoRA file to something diffusers can process. Also I found that we don't...
kohya_controllllite_xl_depth kohya_controllllite_xl_depth_anime kohya_controllllite control models. (control weight: 1) You can suppress the anime style (to some extent) of the Kohya depth model by adding the following terms to the negative prompt. painting, anime, cartoon Stability AI’...
Introduction The prevalence of viral diseases has increased due to the availability of modern diagnostic tests that allow rapid detection of viruses [1]. Viral diseases may additionally be associated with significant morbidity and mortality as is the case with some emerging viral diseases, such as ...
We used inflation rates between 2010 and 2012 in different commodity groups [84] in order to update embedded carbon coefficients for the analysis of household expenditures in 2012. In order to overcome the exclusion of post-purchase emissions in the input-output framework [95], we used a ...
Update: Changed the sample training settings to a learning rate of 2e-4, --timestep_sampling to shift, and --discrete_flow_shift to 7.0. Faster training is expected. If the details of the image are not learned well, try lowering the discete flow shift to around 3.0. However, the train...