Lincoln Stein
a958ae5e29
Merge branch 'main' into feat/use-custom-vaes
2023-03-23 10:32:56 -04:00
psychedelicious
b194180f76
feat(backend): make fast latents method static
2023-03-16 20:03:08 +11:00
psychedelicious
fb30b7d17a
feat(backend): add image_to_dataURL util
2023-03-16 20:03:08 +11:00
Jonathan
44f489d581
Merge branch 'main' into fix-seampaint
2023-03-14 06:19:25 -05:00
blessedcoolant
0a761d7c43
fix(inpaint): Seam painting being broken
2023-03-15 00:00:08 +13:00
Damian Stewart
a0f47aa72e
Merge branch 'main' into main
2023-03-14 11:41:29 +01:00
Lincoln Stein
d840c597b5
fix --png_compression command line argument
...
- The value of png_compression was always 6, despite the value provided to the
--png_compression argument. This fixes the bug.
- It also fixes an inconsistency between the maximum range of png_compression
and the help text.
- Closes #2945
2023-03-14 00:24:05 -04:00
Lincoln Stein
3ca654d256
speculative fix for alternative vaes
2023-03-13 23:27:29 -04:00
jeremy
e0e01f6c50
Reduced Pickle ACE attack surface
...
Prior to this commit, all models would be loaded with the extremely unsafe `torch.load` method, except those with the exact extension `.safetensors`. Even a change in casing (eg. `saFetensors`, `Safetensors`, etc) would cause the file to be loaded with torch.load instead of the much safer `safetensors.toch.load_file`.
If a malicious actor renamed an infected `.ckpt` to something like `.SafeTensors` or `.SAFETENSORS` an unsuspecting user would think they are loading a safe .safetensor, but would in fact be parsing an unsafe pickle file, and executing an attacker's payload. This commit fixes this vulnerability by reversing the loading-method decision logic to only use the unsafe `torch.load` when the file extension is exactly `.ckpt`.
2023-03-13 16:16:30 -04:00
JPPhoto
596ba754b1
Removed seed from get_make_image.
2023-03-13 08:15:46 -05:00
JPPhoto
b980e563b9
Fix bug #2931
2023-03-13 08:11:09 -05:00
Kevin Turner
288cee9611
Merge remote-tracking branch 'origin/main' into feat/preview_predicted_x0
...
# Conflicts:
# invokeai/app/invocations/generate.py
2023-03-12 20:56:02 -07:00
Kyle Schouviller
3ee2798ede
[fix] Get the model again if current model is empty
2023-03-12 22:26:11 -04:00
Fabio 'MrWHO' Torchetti
5c5106c14a
Add keys when non EMA
2023-03-12 16:22:22 -05:00
Fabio 'MrWHO' Torchetti
c367b21c71
Fix issue #2932
2023-03-12 15:40:33 -05:00
Lincoln Stein
6a77634b34
remove unneeded generate initializer routines
2023-03-11 17:14:03 -05:00
Lincoln Stein
8ca91b1774
add restoration services to nodes
2023-03-11 17:00:00 -05:00
Lincoln Stein
3aa1ee1218
restore NSFW checker
2023-03-11 16:16:44 -05:00
Lincoln Stein
c14241436b
move ModelManager initialization into its own module and restore embedding support
2023-03-11 10:56:53 -05:00
Lincoln Stein
d612f11c11
initialize InvokeAIGenerator object with model, not manager
2023-03-11 09:06:46 -05:00
Lincoln Stein
250b0ab182
add seamless tiling support
2023-03-11 08:33:23 -05:00
Lincoln Stein
675dd12b6c
add attention map images to output object
2023-03-11 08:07:01 -05:00
Lincoln Stein
7e76eea059
add embiggen, remove complicated constructor
2023-03-11 07:50:39 -05:00
Lincoln Stein
fe75b95464
Merge branch 'refactor/nodes-on-generator' of github.com:invoke-ai/InvokeAI into refactor/nodes-on-generator
2023-03-10 19:36:40 -05:00
Lincoln Stein
95954188b2
remove factory pattern
...
Factory pattern is now removed. Typical usage of the InvokeAIGenerator is now:
```
from invokeai.backend.generator import (
InvokeAIGeneratorBasicParams,
Txt2Img,
Img2Img,
Inpaint,
)
params = InvokeAIGeneratorBasicParams(
model_name = 'stable-diffusion-1.5',
steps = 30,
scheduler = 'k_lms',
cfg_scale = 8.0,
height = 640,
width = 640
)
print ('=== TXT2IMG TEST ===')
txt2img = Txt2Img(manager, params)
outputs = txt2img.generate(prompt='banana sushi', iterations=2)
for i in outputs:
print(f'image={output.image}, seed={output.seed}, model={output.params.model_name}, hash={output.model_hash}, steps={output.params.steps}')
```
The `params` argument is optional, so if you wish to accept default
parameters and selectively override them, just do this:
```
outputs = Txt2Img(manager).generate(prompt='banana sushi',
steps=50,
scheduler='k_heun',
model_name='stable-diffusion-2.1'
)
```
2023-03-10 19:33:04 -05:00
Jonathan
63f59201f8
Merge branch 'main' into feat/preview_predicted_x0
2023-03-10 12:34:07 -06:00
Jonathan
370e8281b3
Merge branch 'main' into refactor/nodes-on-generator
2023-03-10 12:34:00 -06:00
Lincoln Stein
685df33584
fix bug that caused black images when converting ckpts to diffusers in RAM ( #2914 )
...
Cause of the problem was inadvertent activation of the safety checker.
When conversion occurs on disk, the safety checker is disabled during loading.
However, when converting in RAM, the safety checker was not removed, resulting
in it activating even when user specified --no-nsfw_checker.
This PR fixes the problem by detecting when the caller has requested the InvokeAi
StableDiffusionGeneratorPipeline class to be returned and setting safety checker
to None. Do not do this with diffusers models destined for disk because then they
will be incompatible with the merge script!!
Closes #2836
2023-03-10 18:11:32 +00:00
Kevin Turner
fe6858f2d9
feat: use the predicted denoised image for previews
...
Some schedulers report not only the noisy latents at the current timestep,
but also their estimate so far of what the de-noised latents will be.
It makes for a more legible preview than the noisy latents do.
2023-03-09 20:28:06 -08:00
Lincoln Stein
14c8738a71
fix dangling reference to _model_to_cpu and missing variable model_description
2023-03-09 21:41:45 -05:00
Kevin Turner
1a829bb998
pipeline: remove code for legacy model
2023-03-09 18:15:12 -08:00
Kevin Turner
9d339e94f2
backend..conditioning: remove code for legacy model
2023-03-09 18:15:12 -08:00
Kevin Turner
ad7b1fa6fb
model_manager: model to/from CPU methods are implemented on the Pipeline
2023-03-09 18:15:12 -08:00
Kevin Turner
42355b70c2
fix(Pipeline.debug_latents): fix import for moved utility function
2023-03-09 18:15:12 -08:00
Kevin Turner
faa2558e2f
chore: add new argument to overridden method to match new signature upstream
2023-03-09 18:15:12 -08:00
Kevin Turner
081397737b
typo: docstring spelling fixes
...
looks like they've already been corrected in the upstream copy
2023-03-09 18:15:12 -08:00
Kevin Turner
55d36eaf4f
fix: image_resized_to_grid_as_tensor: reconnect dropped multiple_of argument
2023-03-09 18:15:12 -08:00
Lincoln Stein
c11e823ff3
remove unused _wrap_results
2023-03-09 16:30:06 -05:00
Lincoln Stein
cde0b6ae8d
Merge branch 'main' into refactor/nodes-on-generator
2023-03-09 01:52:45 -05:00
Lincoln Stein
b679a6ba37
model manager defaults to consistent values of device and precision
2023-03-09 01:09:54 -05:00
Lincoln Stein
5d37fa6e36
node-based txt2img working without generate
2023-03-09 00:18:29 -05:00
Damian Stewart
9ee648e0c3
Merge branch 'main' into feat_longer_prompts
2023-03-09 00:13:01 +01:00
Jonathan
2db180d909
Make img2img strength 1 behave the same as txt2img ( #2895 )
...
* Fix img2img and inpainting code so a strength of 1 behaves the same as txt2img.
* Make generated images identical to their txt2img counterparts when strength is 1.
2023-03-08 22:50:16 +01:00
damian
69e2dc0404
update for compel changes
2023-03-08 20:45:01 +01:00
Damian Stewart
a38b75572f
don't log excess tokens as truncated
2023-03-08 20:00:18 +01:00
damian
768e969c90
cleanup and fix kwarg
2023-03-08 18:00:54 +01:00
Damian Stewart
57db66634d
longer prompts wip
2023-03-08 14:25:48 +01:00
Lincoln Stein
87789c1de8
add InvokeAIGenerator and InvokeAIGeneratorFactory classes
2023-03-07 23:52:53 -05:00
Jonathan
c3ff9e6be8
Fixed startup issues with the web UI. ( #2876 )
2023-03-06 18:29:28 -05:00
Lincoln Stein
7c60068388
Merge branch 'main' into bugfix/fix-convert-sd-to-diffusers-error
2023-03-06 08:20:29 -05:00
