Mirrored_Repos/InvokeAI
1
0
Fork 0
mirror of https://github.com/invoke-ai/InvokeAI synced 2024-08-30 20:32:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix preview, inpaint

Browse Source
This commit is contained in:
Sergey Borisov
2023-08-07 21:27:32 +03:00
parent 2539e26c18
commit 1db2c93f75
4 changed files with 58 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

60
invokeai/app/util/step_callback.py
View File

@ -7,6 +7,7 @@ from ...backend.util.util import image_to_dataURL
from ...backend.generator.base import Generator
from ...backend.stable_diffusion import PipelineIntermediateState
from invokeai.app.services.config import InvokeAIAppConfig
from ...backend.model_management.models import BaseModelType
def sample_to_lowres_estimated_image(samples, latent_rgb_factors, smooth_matrix=None):
@ -29,6 +30,7 @@ def stable_diffusion_step_callback(
intermediate_state: PipelineIntermediateState,
node: dict,
source_node_id: str,
base_model: BaseModelType,
):
if context.services.queue.is_canceled(context.graph_execution_state_id):
raise CanceledException
@ -56,23 +58,51 @@ def stable_diffusion_step_callback(
# TODO: only output a preview image when requested
# origingally adapted from code by @erucipe and @keturn here:
# https://discuss.huggingface.co/t/decoding-latents-to-rgb-without-upscaling/23204/7
if base_model in [BaseModelType.StableDiffusionXL, BaseModelType.StableDiffusionXLRefiner]:
sdxl_latent_rgb_factors = torch.tensor(
[
# R G B
[0.3816, 0.4930, 0.5320],
[-0.3753, 0.1631, 0.1739],
[0.1770, 0.3588, -0.2048],
[-0.4350, -0.2644, -0.4289],
],
dtype=sample.dtype,
device=sample.device,
)
# these updated numbers for v1.5 are from @torridgristle
v1_5_latent_rgb_factors = torch.tensor(
[
# R G B
[0.3444, 0.1385, 0.0670], # L1
[0.1247, 0.4027, 0.1494], # L2
[-0.3192, 0.2513, 0.2103], # L3
[-0.1307, -0.1874, -0.7445], # L4
],
dtype=sample.dtype,
device=sample.device,
)
sdxl_smooth_matrix = torch.tensor(
[
# [ 0.0478, 0.1285, 0.0478],
# [ 0.1285, 0.2948, 0.1285],
# [ 0.0478, 0.1285, 0.0478],
[0.0358, 0.0964, 0.0358],
[0.0964, 0.4711, 0.0964],
[0.0358, 0.0964, 0.0358],
],
dtype=sample.dtype,
device=sample.device,
)
image = sample_to_lowres_estimated_image(sample, v1_5_latent_rgb_factors)
image = sample_to_lowres_estimated_image(sample, sdxl_latent_rgb_factors, sdxl_smooth_matrix)
else:
# origingally adapted from code by @erucipe and @keturn here:
# https://discuss.huggingface.co/t/decoding-latents-to-rgb-without-upscaling/23204/7
# these updated numbers for v1.5 are from @torridgristle
v1_5_latent_rgb_factors = torch.tensor(
[
# R G B
[0.3444, 0.1385, 0.0670], # L1
[0.1247, 0.4027, 0.1494], # L2
[-0.3192, 0.2513, 0.2103], # L3
[-0.1307, -0.1874, -0.7445], # L4
],
dtype=sample.dtype,
device=sample.device,
)
image = sample_to_lowres_estimated_image(sample, v1_5_latent_rgb_factors)
(width, height) = image.size
width *= 8