mirror of
https://github.com/invoke-ai/InvokeAI
synced 2024-08-30 20:32:17 +00:00
17 lines
692 B
Python
17 lines
692 B
Python
from typing import Literal
|
|
|
|
from invokeai.backend.util.devices import TorchDevice
|
|
|
|
LATENT_SCALE_FACTOR = 8
|
|
"""
|
|
HACK: Many nodes are currently hard-coded to use a fixed latent scale factor of 8. This is fragile, and will need to
|
|
be addressed if future models use a different latent scale factor. Also, note that there may be places where the scale
|
|
factor is hard-coded to a literal '8' rather than using this constant.
|
|
The ratio of image:latent dimensions is LATENT_SCALE_FACTOR:1, or 8:1.
|
|
"""
|
|
|
|
IMAGE_MODES = Literal["L", "RGB", "RGBA", "CMYK", "YCbCr", "LAB", "HSV", "I", "F"]
|
|
"""A literal type for PIL image modes supported by Invoke"""
|
|
|
|
DEFAULT_PRECISION = TorchDevice.choose_torch_dtype()
|