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

Merge branch 'main' into feat/refactor_generation_backend

Browse Source
This commit is contained in:
Sergey Borisov
2023-08-10 04:32:16 +03:00
parent e98f7eda2e c8fbaf54b6
commit ade78b9591
43 changed files with 1970 additions and 407 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
invokeai/app/api/dependencies.py
View File

@ -55,7 +55,7 @@ logger = InvokeAILogger.getLogger()
class ApiDependencies:
"""Contains and initializes all dependencies for the API"""
invoker: Optional[Invoker] = None
invoker: Invoker
@staticmethod
def initialize(config: InvokeAIAppConfig, event_handler_id: int, logger: Logger = logger):
@ -68,8 +68,9 @@ class ApiDependencies:
output_folder = config.output_path
# TODO: build a file/path manager?
db_location = config.db_path
db_location.parent.mkdir(parents=True, exist_ok=True)
db_path = config.db_path
db_path.parent.mkdir(parents=True, exist_ok=True)
db_location = str(db_path)
graph_execution_manager = SqliteItemStorage[GraphExecutionState](
filename=db_location, table_name="graph_executions"

9
invokeai/app/api/routers/images.py
View File

@ -1,22 +1,20 @@
import io
from typing import Optional
from PIL import Image
from fastapi import Body, HTTPException, Path, Query, Request, Response, UploadFile
from fastapi.responses import FileResponse
from fastapi.routing import APIRouter
from PIL import Image
from pydantic import BaseModel, Field
from pydantic import BaseModel
from invokeai.app.invocations.metadata import ImageMetadata
from invokeai.app.models.image import ImageCategory, ResourceOrigin
from invokeai.app.services.image_record_storage import OffsetPaginatedResults
from invokeai.app.services.item_storage import PaginatedResults
from invokeai.app.services.models.image_record import (
ImageDTO,
ImageRecordChanges,
ImageUrlsDTO,
)
from ..dependencies import ApiDependencies
images_router = APIRouter(prefix="/v1/images", tags=["images"])
@ -152,8 +150,9 @@ async def get_image_metadata(
raise HTTPException(status_code=404)
@images_router.get(
@images_router.api_route(
"/i/{image_name}/full",
methods=["GET", "HEAD"],
operation_id="get_image_full",
response_class=Response,
responses={

144
invokeai/app/invocations/image.py
View File

@ -4,6 +4,7 @@ from typing import Literal, Optional
import cv2
import numpy
import cv2
from PIL import Image, ImageFilter, ImageOps, ImageChops
from pydantic import Field
from pathlib import Path
@ -502,7 +503,7 @@ class ImageLerpInvocation(BaseInvocation, PILInvocationConfig):
image = context.services.images.get_pil_image(self.image.image_name)
image_arr = numpy.asarray(image, dtype=numpy.float32) / 255
image_arr = image_arr * (self.max - self.min) + self.max
image_arr = image_arr * (self.max - self.min) + self.min
lerp_image = Image.fromarray(numpy.uint8(image_arr))
@ -653,6 +654,7 @@ class ImageWatermarkInvocation(BaseInvocation, PILInvocationConfig):
height=image_dto.height,
)
class MaskEdgeInvocation(BaseInvocation, PILInvocationConfig):
"""Applies an edge mask to an image"""
@ -702,6 +704,7 @@ class MaskEdgeInvocation(BaseInvocation, PILInvocationConfig):
height=image_dto.height,
)
class ColorCorrectInvocation(BaseInvocation, PILInvocationConfig):
type: Literal["color_correct"] = "color_correct"
@ -817,3 +820,142 @@ class ColorCorrectInvocation(BaseInvocation, PILInvocationConfig):
height=image_dto.height,
)
class ImageHueAdjustmentInvocation(BaseInvocation):
"""Adjusts the Hue of an image."""
# fmt: off
type: Literal["img_hue_adjust"] = "img_hue_adjust"
# Inputs
image: ImageField = Field(default=None, description="The image to adjust")
hue: int = Field(default=0, description="The degrees by which to rotate the hue, 0-360")
# fmt: on
def invoke(self, context: InvocationContext) -> ImageOutput:
pil_image = context.services.images.get_pil_image(self.image.image_name)
# Convert image to HSV color space
hsv_image = numpy.array(pil_image.convert("HSV"))
# Convert hue from 0..360 to 0..256
hue = int(256 * ((self.hue % 360) / 360))
# Increment each hue and wrap around at 255
hsv_image[:, :, 0] = (hsv_image[:, :, 0] + hue) % 256
# Convert back to PIL format and to original color mode
pil_image = Image.fromarray(hsv_image, mode="HSV").convert("RGBA")
image_dto = context.services.images.create(
image=pil_image,
image_origin=ResourceOrigin.INTERNAL,
image_category=ImageCategory.GENERAL,
node_id=self.id,
is_intermediate=self.is_intermediate,
session_id=context.graph_execution_state_id,
)
return ImageOutput(
image=ImageField(
image_name=image_dto.image_name,
),
width=image_dto.width,
height=image_dto.height,
)
class ImageLuminosityAdjustmentInvocation(BaseInvocation):
"""Adjusts the Luminosity (Value) of an image."""
# fmt: off
type: Literal["img_luminosity_adjust"] = "img_luminosity_adjust"
# Inputs
image: ImageField = Field(default=None, description="The image to adjust")
luminosity: float = Field(default=1.0, ge=0, le=1, description="The factor by which to adjust the luminosity (value)")
# fmt: on
def invoke(self, context: InvocationContext) -> ImageOutput:
pil_image = context.services.images.get_pil_image(self.image.image_name)
# Convert PIL image to OpenCV format (numpy array), note color channel
# ordering is changed from RGB to BGR
image = numpy.array(pil_image.convert("RGB"))[:, :, ::-1]
# Convert image to HSV color space
hsv_image = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
# Adjust the luminosity (value)
hsv_image[:, :, 2] = numpy.clip(hsv_image[:, :, 2] * self.luminosity, 0, 255)
# Convert image back to BGR color space
image = cv2.cvtColor(hsv_image, cv2.COLOR_HSV2BGR)
# Convert back to PIL format and to original color mode
pil_image = Image.fromarray(image[:, :, ::-1], "RGB").convert("RGBA")
image_dto = context.services.images.create(
image=pil_image,
image_origin=ResourceOrigin.INTERNAL,
image_category=ImageCategory.GENERAL,
node_id=self.id,
is_intermediate=self.is_intermediate,
session_id=context.graph_execution_state_id,
)
return ImageOutput(
image=ImageField(
image_name=image_dto.image_name,
),
width=image_dto.width,
height=image_dto.height,
)
class ImageSaturationAdjustmentInvocation(BaseInvocation):
"""Adjusts the Saturation of an image."""
# fmt: off
type: Literal["img_saturation_adjust"] = "img_saturation_adjust"
# Inputs
image: ImageField = Field(default=None, description="The image to adjust")
saturation: float = Field(default=1.0, ge=0, le=1, description="The factor by which to adjust the saturation")
# fmt: on
def invoke(self, context: InvocationContext) -> ImageOutput:
pil_image = context.services.images.get_pil_image(self.image.image_name)
# Convert PIL image to OpenCV format (numpy array), note color channel
# ordering is changed from RGB to BGR
image = numpy.array(pil_image.convert("RGB"))[:, :, ::-1]
# Convert image to HSV color space
hsv_image = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
# Adjust the saturation
hsv_image[:, :, 1] = numpy.clip(hsv_image[:, :, 1] * self.saturation, 0, 255)
# Convert image back to BGR color space
image = cv2.cvtColor(hsv_image, cv2.COLOR_HSV2BGR)
# Convert back to PIL format and to original color mode
pil_image = Image.fromarray(image[:, :, ::-1], "RGB").convert("RGBA")
image_dto = context.services.images.create(
image=pil_image,
image_origin=ResourceOrigin.INTERNAL,
image_category=ImageCategory.GENERAL,
node_id=self.id,
is_intermediate=self.is_intermediate,
session_id=context.graph_execution_state_id,
)
return ImageOutput(
image=ImageField(
image_name=image_dto.image_name,
),
width=image_dto.width,
height=image_dto.height,
)

16
invokeai/app/invocations/latent.py
View File

@ -5,15 +5,26 @@ from typing import List, Literal, Optional, Union
import einops
import torch
from diffusers import ControlNetModel
from diffusers.image_processor import VaeImageProcessor
from diffusers.models.attention_processor import (
AttnProcessor2_0,
LoRAAttnProcessor2_0,
LoRAXFormersAttnProcessor,
XFormersAttnProcessor,
)
from diffusers.schedulers import SchedulerMixin as Scheduler
from pydantic import BaseModel, Field, validator
from invokeai.app.invocations.metadata import CoreMetadata
from invokeai.app.util.controlnet_utils import prepare_control_image
from invokeai.app.util.step_callback import stable_diffusion_step_callback
from invokeai.backend.model_management.models import ModelType, SilenceWarnings
from .baseinvocation import BaseInvocation, BaseInvocationOutput, InvocationConfig, InvocationContext
from .compel import ConditioningField
from .controlnet_image_processors import ControlField
from .image import ImageOutput
from .model import ModelInfo, UNetField, VaeField
from ..models.image import ImageCategory, ImageField, ResourceOrigin
from ...backend.model_management import ModelPatcher
from ...backend.stable_diffusion import PipelineIntermediateState
from ...backend.stable_diffusion.diffusers_pipeline import (
@ -239,7 +250,6 @@ class TextToLatentsInvocation(BaseInvocation):
safety_checker=None,
feature_extractor=None,
requires_safety_checker=False,
precision="float16" if unet.dtype == torch.float16 else "float32",
)
def prep_control_data(

28
invokeai/app/services/config.py
View File

@ -24,11 +24,10 @@ InvokeAI:
sequential_guidance: false
precision: float16
max_cache_size: 6
max_vram_cache_size: 2.7
max_vram_cache_size: 0.5
always_use_cpu: false
free_gpu_mem: false
Features:
restore: true
esrgan: true
patchmatch: true
internet_available: true
@ -165,7 +164,7 @@ import pydoc
import os
import sys
from argparse import ArgumentParser
from omegaconf import OmegaConf, DictConfig
from omegaconf import OmegaConf, DictConfig, ListConfig
from pathlib import Path
from pydantic import BaseSettings, Field, parse_obj_as
from typing import ClassVar, Dict, List, Set, Literal, Union, get_origin, get_type_hints, get_args
@ -173,6 +172,7 @@ from typing import ClassVar, Dict, List, Set, Literal, Union, get_origin, get_ty
INIT_FILE = Path("invokeai.yaml")
DB_FILE = Path("invokeai.db")
LEGACY_INIT_FILE = Path("invokeai.init")
DEFAULT_MAX_VRAM = 0.5
class InvokeAISettings(BaseSettings):
@ -189,7 +189,12 @@ class InvokeAISettings(BaseSettings):
opt = parser.parse_args(argv)
for name in self.__fields__:
if name not in self._excluded():
setattr(self, name, getattr(opt, name))
value = getattr(opt, name)
if isinstance(value, ListConfig):
value = list(value)
elif isinstance(value, DictConfig):
value = dict(value)
setattr(self, name, value)
def to_yaml(self) -> str:
"""
@ -282,14 +287,10 @@ class InvokeAISettings(BaseSettings):
return [
"type",
"initconf",
"gpu_mem_reserved",
"max_loaded_models",
"version",
"from_file",
"model",
"restore",
"root",
"nsfw_checker",
]
class Config:
@ -388,15 +389,11 @@ class InvokeAIAppConfig(InvokeAISettings):
internet_available : bool = Field(default=True, description="If true, attempt to download models on the fly; otherwise only use local models", category='Features')
log_tokenization : bool = Field(default=False, description="Enable logging of parsed prompt tokens.", category='Features')
patchmatch : bool = Field(default=True, description="Enable/disable patchmatch inpaint code", category='Features')
restore : bool = Field(default=True, description="Enable/disable face restoration code (DEPRECATED)", category='DEPRECATED')
always_use_cpu : bool = Field(default=False, description="If true, use the CPU for rendering even if a GPU is available.", category='Memory/Performance')
free_gpu_mem : bool = Field(default=False, description="If true, purge model from GPU after each generation.", category='Memory/Performance')
max_loaded_models : int = Field(default=3, gt=0, description="(DEPRECATED: use max_cache_size) Maximum number of models to keep in memory for rapid switching", category='DEPRECATED')
max_cache_size : float = Field(default=6.0, gt=0, description="Maximum memory amount used by model cache for rapid switching", category='Memory/Performance')
max_vram_cache_size : float = Field(default=2.75, ge=0, description="Amount of VRAM reserved for model storage", category='Memory/Performance')
gpu_mem_reserved : float = Field(default=2.75, ge=0, description="DEPRECATED: use max_vram_cache_size. Amount of VRAM reserved for model storage", category='DEPRECATED')
nsfw_checker : bool = Field(default=True, description="DEPRECATED: use Web settings to enable/disable", category='DEPRECATED')
precision : Literal[tuple(['auto','float16','float32','autocast'])] = Field(default='auto',description='Floating point precision', category='Memory/Performance')
sequential_guidance : bool = Field(default=False, description="Whether to calculate guidance in serial instead of in parallel, lowering memory requirements", category='Memory/Performance')
xformers_enabled : bool = Field(default=True, description="Enable/disable memory-efficient attention", category='Memory/Performance')
@ -414,9 +411,7 @@ class InvokeAIAppConfig(InvokeAISettings):
outdir : Path = Field(default='outputs', description='Default folder for output images', category='Paths')
from_file : Path = Field(default=None, description='Take command input from the indicated file (command-line client only)', category='Paths')
use_memory_db : bool = Field(default=False, description='Use in-memory database for storing image metadata', category='Paths')
ignore_missing_core_models : bool = Field(default=False, description='Ignore missing models in models/core/convert')
model : str = Field(default='stable-diffusion-1.5', description='Initial model name', category='Models')
ignore_missing_core_models : bool = Field(default=False, description='Ignore missing models in models/core/convert', category='Features')
log_handlers : List[str] = Field(default=["console"], description='Log handler. Valid options are "console", "file=<path>", "syslog=path|address:host:port", "http=<url>"', category="Logging")
# note - would be better to read the log_format values from logging.py, but this creates circular dependencies issues
@ -426,6 +421,9 @@ class InvokeAIAppConfig(InvokeAISettings):
version : bool = Field(default=False, description="Show InvokeAI version and exit", category="Other")
# fmt: on
class Config:
validate_assignment = True
def parse_args(self, argv: List[str] = None, conf: DictConfig = None, clobber=False):
"""
Update settings with contents of init file, environment, and

23
invokeai/app/services/model_manager_service.py
View File

@ -3,9 +3,10 @@
from __future__ import annotations
from abc import ABC, abstractmethod
from logging import Logger
from pathlib import Path
from pydantic import Field
from typing import Optional, Union, Callable, List, Tuple, TYPE_CHECKING
from typing import Literal, Optional, Union, Callable, List, Tuple, TYPE_CHECKING
from types import ModuleType
from invokeai.backend.model_management import (
@ -193,7 +194,7 @@ class ModelManagerServiceBase(ABC):
self,
model_name: str,
base_model: BaseModelType,
model_type: Union[ModelType.Main, ModelType.Vae],
model_type: Literal[ModelType.Main, ModelType.Vae],
) -> AddModelResult:
"""
Convert a checkpoint file into a diffusers folder, deleting the cached
@ -292,7 +293,7 @@ class ModelManagerService(ModelManagerServiceBase):
def __init__(
self,
config: InvokeAIAppConfig,
logger: ModuleType,
logger: Logger,
):
"""
Initialize with the path to the models.yaml config file.
@ -396,7 +397,7 @@ class ModelManagerService(ModelManagerServiceBase):
model_type,
)
def model_info(self, model_name: str, base_model: BaseModelType, model_type: ModelType) -> dict:
def model_info(self, model_name: str, base_model: BaseModelType, model_type: ModelType) -> Union[dict, None]:
"""
Given a model name returns a dict-like (OmegaConf) object describing it.
"""
@ -416,7 +417,7 @@ class ModelManagerService(ModelManagerServiceBase):
"""
return self.mgr.list_models(base_model, model_type)
def list_model(self, model_name: str, base_model: BaseModelType, model_type: ModelType) -> dict:
def list_model(self, model_name: str, base_model: BaseModelType, model_type: ModelType) -> Union[dict, None]:
"""
Return information about the model using the same format as list_models()
"""
@ -429,7 +430,7 @@ class ModelManagerService(ModelManagerServiceBase):
model_type: ModelType,
model_attributes: dict,
clobber: bool = False,
) -> None:
) -> AddModelResult:
"""
Update the named model with a dictionary of attributes. Will fail with an
assertion error if the name already exists. Pass clobber=True to overwrite.
@ -478,7 +479,7 @@ class ModelManagerService(ModelManagerServiceBase):
self,
model_name: str,
base_model: BaseModelType,
model_type: Union[ModelType.Main, ModelType.Vae],
model_type: Literal[ModelType.Main, ModelType.Vae],
convert_dest_directory: Optional[Path] = Field(
default=None, description="Optional directory location for merged model"
),
@ -573,9 +574,9 @@ class ModelManagerService(ModelManagerServiceBase):
default=None, description="Base model shared by all models to be merged"
),
merged_model_name: str = Field(default=None, description="Name of destination model after merging"),
alpha: Optional[float] = 0.5,
alpha: float = 0.5,
interp: Optional[MergeInterpolationMethod] = None,
force: Optional[bool] = False,
force: bool = False,
merge_dest_directory: Optional[Path] = Field(
default=None, description="Optional directory location for merged model"
),
@ -633,8 +634,8 @@ class ModelManagerService(ModelManagerServiceBase):
model_name: str,
base_model: BaseModelType,
model_type: ModelType,
new_name: str = None,
new_base: BaseModelType = None,
new_name: Optional[str] = None,
new_base: Optional[BaseModelType] = None,
):
"""
Rename the indicated model. Can provide a new name and/or a new base.