- the functionality to automatically import and run legacy checkpoint
files in a designated folder has been removed from the backend but there
are vestiges of the code remaining in the frontend that are causing
crashes.
- This fixes the problem.
- Closes#3075
This PR introduces a new set of ModelManager methods that enables you to
retrieve the individual parts of a stable diffusion pipeline model,
including the vae, text_encoder, unet, tokenizer, etc.
To use:
```
from invokeai.backend import ModelManager
manager = ModelManager('/path/to/models.yaml')
# get the VAE
vae = manager.get_model_vae('stable-diffusion-1.5')
# get the unet
unet = manager.get_model_unet('stable-diffusion-1.5')
# get the tokenizer
tokenizer = manager.get_model_tokenizer('stable-diffusion-1.5')
# etc etc
feature_extractor = manager.get_model_feature_extractor('stable-diffusion-1.5')
scheduler = manager.get_model_scheduler('stable-diffusion-1.5')
text_encoder = manager.get_model_text_encoder('stable-diffusion-1.5')
# if no model provided, then defaults to the one currently in GPU, if any
vae = manager.get_model_vae()
```
- Compensatory change to the CLI that prevents it from crashing
when it tries to import a model.
- Bug introduced when the "convert" option removed from the model
manager.
* Add latents nodes.
* Fix iteration expansion.
* Add collection generator nodes, math nodes.
* Add noise node.
* Add some graph debug commands to the CLI.
* Fix negative id linking in CLI.
* Fix a CLI bug with multiple links per node.
- New method is ModelManager.get_sub_model(model_name:str,model_part:SDModelComponent)
To use:
```
from invokeai.backend import ModelManager, SDModelComponent as sdmc
manager = ModelManager('/path/to/models.yaml')
vae = manager.get_sub_model('stable-diffusion-1.5', sdmc.vae)
```
The typo accidentally did not affect functionality; when `query==""`, it
`search()`ed but found everything due to empty query, then paginated
results, so it worked the same as `list()`.
Still fix it
currently if users input eg `happy (camper:0.3)` it gets parsed
incorrectly, which causes crashes if it's in the negative prompt. bump
to compel 1.0.5 fixes the parser to avoid this (note the weight is
parsed as plain text, it's not converted to proper invoke syntax)
- This PR adds support for embedding files that contain a single key
"emb_params". The only example I know of this format is the
"EasyNegative" embedding on HuggingFace, but there are certainly others.
- This PR also adds support for loading embedding files that have been
saved in safetensors format.
- It also cleans up the code so that the logic of probing for and
selecting the right format parser is clear.
- This is the same as #3045, which is on the 2.3 branch.
- Commands, invocations and their parameters will now autocomplete using
introspection.
- Two types of parameter *arguments* will also autocomplete:
- --sampler_name will autocomplete the scheduler name
- --model will autocomplete the model name
- There don't seem to be commands for reading/writing image files yet,
so path autocompletion is not implemented
A long-standing issue with importing legacy checkpoints (both ckpt and
safetensors) is that the user has to identify the correct config file,
either by providing its path or by selecting which type of model the
checkpoint is (e.g. "v1 inpainting"). In addition, some users wish to
provide custom VAEs for use with the model. Currently this is done in
the WebUI by importing the model, editing it, and then typing in the
path to the VAE.
## Model configuration file selection
To improve the user experience, the model manager's `heuristic_import()`
method has been enhanced as follows:
1. When initially called, the caller can pass a config file path, in
which case it will be used.
2. If no config file provided, the method looks for a .yaml file in the
same directory as the model which bears the same basename. e.g.
```
my-new-model.safetensors
my-new-model.yaml
```
The yaml file is then used as the configuration file for importation and
conversion.
3. If no such file is found, then the method opens up the checkpoint and
probes it to determine whether it is V1, V1-inpaint or V2. If it is a V1
format, then the appropriate v1-inference.yaml config file is used.
Unfortunately there are two V2 variants that cannot be distinguished by
introspection.
4. If the probe algorithm is unable to determine the model type, then
its last-ditch effort is to execute an optional callback function that
can be provided by the caller. This callback, named
`config_file_callback` receives the path to the legacy checkpoint and
returns the path to the config file to use. The CLI uses to put up a
multiple choice prompt to the user. The WebUI **could** use this to
prompt the user to choose from a radio-button selection.
5. If the config file cannot be determined, then the import is
abandoned.
## Custom VAE Selection
The user can attach a custom VAE to the imported and converted model by
copying the desired VAE into the same directory as the file to be
imported, and giving it the same basename. E.g.:
```
my-new-model.safetensors
my-new-model.vae.pt
```
For this to work, the VAE must end with ".vae.pt", ".vae.ckpt", or
".vae.safetensors". The indicated VAE will be converted into diffusers
format and stored with the converted models file, so the ".pt" file can
be deleted after conversion.
No facility is currently provided to swap a diffusers VAE at import
time, but this can be done after the fact using the WebUI and CLI's
model editing functions.
Note that this is the same fix that was applied to the 2.3 branch in
#3043 . This applies to `main`.
## Enable the on-the-fly conversion of models based on SD 2.0/2.1 into
diffusers
This commit fixes bugs related to the on-the-fly conversion and loading
of legacy checkpoint models built on SD-2.0 base.
- When legacy checkpoints built on SD-2.0 models were converted
on-the-fly using --ckpt_convert, generation would crash with a precision
incompatibility error. This problem has been found and fixed.
