A new infill method, **solid:** solid color. currently using middle
gray.
Fixes#2417
It seems like the runwayml inpainting model specifically expects those
masked areas to be blanked out like this.
I haven't tried the SD 2.0 inpainting model with it yet.
Otherwise the model seems too reluctant to change these areas, even
though the mask channel should allow it to.
This makes the solid infill method proposed by #2441 less necessary,
though I think there's still a place for an infill method that is faster
than patchmatch and more predictable than tiles.
Even with #2441, this PR is still useful because it influences all areas
to be painted, not just the infill area.
Fixes#2417
- fixes a spurious "unknown model name" error when trying to edit the
short name of an existing model.
- relaxes naming requirements to include the ':' and '/' characters
in model names
* Update --hires_fix
Change `--hires_fix` to calculate initial width and height based on the model's resolution (if available) and with a minimum size.
The original textual inversion script in scripts is now superseded. The
replacement can be found in ldm/invoke/textual_inversion.py and is a
merging of the command line and front end scripts. After running `pip
install -e .` there will be a `textual_inversion` command on your path.
You can activate the front end this way:
`textual_inversion -gui`
Adds double-click to reset canvas view to 100%.
- Adds hook to manage single and double clicks
- Single Click `Reset Canvas View` --> scale to fit, no change to
current behaviour
- Double Click `Reset Canvas View` --> set scale to 1
Testing suggests that the diffusers versions of Waifu-1.4 anything-v4.0
require the `sd-vae-ft-mse` to generate decent images, so the
appropriate arguments have been added to the initial model file.
- Model merging and textual inversion scripts have been moved into
`ldm/invoke`, which allows them to be installed properly by
pyproject.toml.
- As part of the pyproject install, the .py suffix is removed from the
command. I.e. use `invoke`, `configure_invokeai`, `merge_models` and
`textual_inversion`.
- GUI versions are activated by adding `--gui` to the command. Without
this, you get a classical argv-based command. Example: `merge_models
--gui`
- Fixed up the launcher scripts to accommodate new naming scheme.
- Keyboard behavior of the GUI front ends has been improved. You can now
use up and down arrow to move from field to field, in addition to <tab>
and ctrl-N/ctrl-P
So far the slider component was unable to take typed input due to a
bunch of issues that were a pain to solve. This PR fixes it.
Things to test:
- Moving the slider also updates the value in the input text box.
- Input text box next to slider can be changed in two ways: If you type
a manual value, the slider will be updated when you lose focus from the
input box. If you use the stepper icons to update the values, the slider
should update immediately.
- Make sure the reset buttons next to the slider are updating correctly
and make sure this updates both the slider and the input box values.
- Brush Size slider -> make sure the hotkeys are updating the input box
too.
