Copied into InvokeAI since IP-Adapter repo is not a package. Is there a better way to do this for non-packaged Python code while still keeping InvokeAI install easy?
## What type of PR is this? (check all applicable)
- [ ] Refactor
- [x] Feature
- [ ] Bug Fix
- [ ] Optimization
- [ ] Documentation Update
- [ ] Community Node Submission
## Description
- Workflows are saved to image files directly
- Image-outputting nodes have an `Embed Workflow` checkbox which, if
enabled, saves the workflow
- `BaseInvocation` now has an `workflow: Optional[str]` field, so all
nodes automatically have the field (but again only image-outputting
nodes display this in UI)
- If this field is enabled, when the graph is created, the workflow is
stringified and set in this field
- Nodes should add `workflow=self.workflow` when they save their output
image to have the workflow written to the image
- Uploads now have their metadata retained so that you can upload
somebody else's image and have access to that workflow
- Graphs are no longer saved to images, workflows replace them
### TODO
- Images created in the linear UI do not have a workflow saved yet. Need
to write a function to build a workflow around the linear UI graph when
using linear tabs. Unfortunately it will not have the nice positioning
and size data the node editor gives you when you save a workflow...
we'll have to figure out how to handle this.
## Related Tickets & Documents
<!--
For pull requests that relate or close an issue, please include them
below.
For example having the text: "closes #1234" would connect the current
pull
request to issue 1234. And when we merge the pull request, Github will
automatically close the issue.
-->
- Related Issue #
- Closes #
## QA Instructions, Screenshots, Recordings
<!--
Please provide steps on how to test changes, any hardware or
software specifications as well as any other pertinent information.
-->
All invocation metadata (type, title, tags and category) are now defined in decorators.
The decorators add the `type: Literal["invocation_type"]: "invocation_type"` field to the invocation.
Category is a new invocation metadata, but it is not used by the frontend just yet.
- `@invocation()` decorator for invocations
```py
@invocation(
"sdxl_compel_prompt",
title="SDXL Prompt",
tags=["sdxl", "compel", "prompt"],
category="conditioning",
)
class SDXLCompelPromptInvocation(BaseInvocation, SDXLPromptInvocationBase):
...
```
- `@invocation_output()` decorator for invocation outputs
```py
@invocation_output("clip_skip_output")
class ClipSkipInvocationOutput(BaseInvocationOutput):
...
```
- update invocation docs
- add category to decorator
- regen frontend types
## What type of PR is this? (check all applicable)
- [x] Feature
- [x] Bug Fix
## Have you discussed this change with the InvokeAI team?
- [x] Yes
## Description
- Keep Boards Modal open by default.
- Combine Coherence and Mask settings under Compositing
- Auto Change Dimensions based on model type (option)
- Size resets are now model dependent
- Add Set Control Image Height & Width to Width and Height option.
- Fix numerous color & spacing issues (especially those pertaining to
sliders being too close to the bottom)
- Add Lock Ratio Option
## What type of PR is this? (check all applicable)
- [ ] Refactor
- [ ] Feature
- [x] Bug Fix
- [ ] Optimization
- [ ] Documentation Update
- [ ] Community Node Submission
## Have you discussed this change with the InvokeAI team?
- [x] Yes
- [ ] No, because:
## Have you updated all relevant documentation?
- [ ] Yes
- [ ] No
## Description
## Related Tickets & Documents
## QA Instructions, Screenshots, Recordings
## Added/updated tests?
- [ ] Yes
- [ ] No : _please replace this line with details on why tests
have not been included_
