# Description
More update/changes to flet wedui, still WIP.
added re-orderable layers panel.
added messages to bottom panel.
changed appbar to row.
# Checklist:
- [x] I have changed the base branch to `dev`
- [x] I have performed a self-review of my own code
- [x] I have commented my code in hard-to-understand areas
- [ ] I have made corresponding changes to the documentation
# Description
Frame generation is being done in chunks of the number of samples
chosen, so when you needed 120 frames but set 200 samples, it would
generate 200 frames.
* Added a break to stop frames from generating once total_frames is
reached
# Checklist:
- [x] I have changed the base branch to `dev`
- [x] I have performed a self-review of my own code
- [ ] I have commented my code in hard-to-understand areas
- [ ] I have made corresponding changes to the documentation
- Fixed issue with new ldm folder requiring the personalization_config
to be set even if empty.
- Added shutup as a dependency to shutup python warnings for good.
# Description
made factory functions to parse config and generate settings page.
# Checklist:
- [x] I have changed the base branch to `dev`
- [x] I have performed a self-review of my own code
- [x] I have commented my code in hard-to-understand areas
- [ ] I have made corresponding changes to the documentation
Initial support for the Stable Diffusion 2.0 model.
As of right now changing the `default_model_config` option on the
`webui_streamli.yaml` to point to the `v2-inference.yaml` file and
downloading manually the Stable Diffusion v2.0 model and placing it on
the `models/custom` folder will let you load it and run it but the
colors are still messed up, needs more work on it to fix this issue.
Fixed the script that convert ckpt models to diffusers not working with
some models if they had invalid or broken keys, these keys are now
ignored so we can still convert the model.
Fixed the script that convert ckpt models to diffusers not working with some models if they had invalid or broken keys, these keys are now ignored so we can still convert the model.
