Merge branch 'main' into lstein/feat/multi-gpu

Author: lstein

invokeai/app/invocations/controlnet_image_processors.py

@@ -165,13 +165,13 @@ def invoke(self, context: InvocationContext) -> ImageOutput:
     title="Canny Processor",
     tags=["controlnet", "canny"],
     category="controlnet",
-    version="1.3.2",
+    version="1.3.3",
 )
 class CannyImageProcessorInvocation(ImageProcessorInvocation):
     """Canny edge detection for ControlNet"""
 
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
     low_threshold: int = InputField(
         default=100, ge=0, le=255, description="The low threshold of the Canny pixel gradient (0-255)"
     )
@@ -199,13 +199,13 @@ def run_processor(self, image: Image.Image) -> Image.Image:
     title="HED (softedge) Processor",
     tags=["controlnet", "hed", "softedge"],
     category="controlnet",
-    version="1.2.2",
+    version="1.2.3",
 )
 class HedImageProcessorInvocation(ImageProcessorInvocation):
     """Applies HED edge detection to image"""
 
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
     # safe not supported in controlnet_aux v0.0.3
     # safe: bool = InputField(default=False, description=FieldDescriptions.safe_mode)
     scribble: bool = InputField(default=False, description=FieldDescriptions.scribble_mode)
@@ -228,13 +228,13 @@ def run_processor(self, image: Image.Image) -> Image.Image:
     title="Lineart Processor",
     tags=["controlnet", "lineart"],
     category="controlnet",
-    version="1.2.2",
+    version="1.2.3",
 )
 class LineartImageProcessorInvocation(ImageProcessorInvocation):
     """Applies line art processing to image"""
 
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
     coarse: bool = InputField(default=False, description="Whether to use coarse mode")
 
     def run_processor(self, image: Image.Image) -> Image.Image:
@@ -250,13 +250,13 @@ def run_processor(self, image: Image.Image) -> Image.Image:
     title="Lineart Anime Processor",
     tags=["controlnet", "lineart", "anime"],
     category="controlnet",
-    version="1.2.2",
+    version="1.2.3",
 )
 class LineartAnimeImageProcessorInvocation(ImageProcessorInvocation):
     """Applies line art anime processing to image"""
 
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
     def run_processor(self, image: Image.Image) -> Image.Image:
         processor = LineartAnimeProcessor()
@@ -273,15 +273,15 @@ def run_processor(self, image: Image.Image) -> Image.Image:
     title="Midas Depth Processor",
     tags=["controlnet", "midas"],
     category="controlnet",
-    version="1.2.3",
+    version="1.2.4",
 )
 class MidasDepthImageProcessorInvocation(ImageProcessorInvocation):
     """Applies Midas depth processing to image"""
 
     a_mult: float = InputField(default=2.0, ge=0, description="Midas parameter `a_mult` (a = a_mult * PI)")
     bg_th: float = InputField(default=0.1, ge=0, description="Midas parameter `bg_th`")
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
     # depth_and_normal not supported in controlnet_aux v0.0.3
     # depth_and_normal: bool = InputField(default=False, description="whether to use depth and normal mode")
 
@@ -304,13 +304,13 @@ def run_processor(self, image):
     title="Normal BAE Processor",
     tags=["controlnet"],
     category="controlnet",
-    version="1.2.2",
+    version="1.2.3",
 )
 class NormalbaeImageProcessorInvocation(ImageProcessorInvocation):
     """Applies NormalBae processing to image"""
 
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
     def run_processor(self, image):
         normalbae_processor = NormalBaeDetector.from_pretrained("lllyasviel/Annotators")
@@ -321,13 +321,13 @@ def run_processor(self, image):
 
 
 @invocation(
-    "mlsd_image_processor", title="MLSD Processor", tags=["controlnet", "mlsd"], category="controlnet", version="1.2.2"
+    "mlsd_image_processor", title="MLSD Processor", tags=["controlnet", "mlsd"], category="controlnet", version="1.2.3"
 )
 class MlsdImageProcessorInvocation(ImageProcessorInvocation):
     """Applies MLSD processing to image"""
 
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
     thr_v: float = InputField(default=0.1, ge=0, description="MLSD parameter `thr_v`")
     thr_d: float = InputField(default=0.1, ge=0, description="MLSD parameter `thr_d`")
 
@@ -344,13 +344,13 @@ def run_processor(self, image):
 
 
 @invocation(
-    "pidi_image_processor", title="PIDI Processor", tags=["controlnet", "pidi"], category="controlnet", version="1.2.2"
+    "pidi_image_processor", title="PIDI Processor", tags=["controlnet", "pidi"], category="controlnet", version="1.2.3"
 )
 class PidiImageProcessorInvocation(ImageProcessorInvocation):
     """Applies PIDI processing to image"""
 
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
     safe: bool = InputField(default=False, description=FieldDescriptions.safe_mode)
     scribble: bool = InputField(default=False, description=FieldDescriptions.scribble_mode)
 
@@ -371,13 +371,13 @@ def run_processor(self, image):
     title="Content Shuffle Processor",
     tags=["controlnet", "contentshuffle"],
     category="controlnet",
-    version="1.2.2",
+    version="1.2.3",
 )
 class ContentShuffleImageProcessorInvocation(ImageProcessorInvocation):
     """Applies content shuffle processing to image"""
 
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
     h: int = InputField(default=512, ge=0, description="Content shuffle `h` parameter")
     w: int = InputField(default=512, ge=0, description="Content shuffle `w` parameter")
     f: int = InputField(default=256, ge=0, description="Content shuffle `f` parameter")
@@ -401,7 +401,7 @@ def run_processor(self, image):
     title="Zoe (Depth) Processor",
     tags=["controlnet", "zoe", "depth"],
     category="controlnet",
-    version="1.2.2",
+    version="1.2.3",
 )
 class ZoeDepthImageProcessorInvocation(ImageProcessorInvocation):
     """Applies Zoe depth processing to image"""
@@ -417,15 +417,15 @@ def run_processor(self, image):
     title="Mediapipe Face Processor",
     tags=["controlnet", "mediapipe", "face"],
     category="controlnet",
-    version="1.2.3",
+    version="1.2.4",
 )
 class MediapipeFaceProcessorInvocation(ImageProcessorInvocation):
     """Applies mediapipe face processing to image"""
 
     max_faces: int = InputField(default=1, ge=1, description="Maximum number of faces to detect")
     min_confidence: float = InputField(default=0.5, ge=0, le=1, description="Minimum confidence for face detection")
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
     def run_processor(self, image):
         mediapipe_face_processor = MediapipeFaceDetector()
@@ -444,16 +444,16 @@ def run_processor(self, image):
     title="Leres (Depth) Processor",
     tags=["controlnet", "leres", "depth"],
     category="controlnet",
-    version="1.2.2",
+    version="1.2.3",
 )
 class LeresImageProcessorInvocation(ImageProcessorInvocation):
     """Applies leres processing to image"""
 
     thr_a: float = InputField(default=0, description="Leres parameter `thr_a`")
     thr_b: float = InputField(default=0, description="Leres parameter `thr_b`")
     boost: bool = InputField(default=False, description="Whether to use boost mode")
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
     def run_processor(self, image):
         leres_processor = LeresDetector.from_pretrained("lllyasviel/Annotators")
@@ -473,7 +473,7 @@ def run_processor(self, image):
     title="Tile Resample Processor",
     tags=["controlnet", "tile"],
     category="controlnet",
-    version="1.2.2",
+    version="1.2.3",
 )
 class TileResamplerProcessorInvocation(ImageProcessorInvocation):
     """Tile resampler processor"""
@@ -513,13 +513,13 @@ def run_processor(self, img):
     title="Segment Anything Processor",
     tags=["controlnet", "segmentanything"],
     category="controlnet",
-    version="1.2.3",
+    version="1.2.4",
 )
 class SegmentAnythingProcessorInvocation(ImageProcessorInvocation):
     """Applies segment anything processing to image"""
 
-    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    detect_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
     def run_processor(self, image):
         # segment_anything_processor = SamDetector.from_pretrained("ybelkada/segment-anything", subfolder="checkpoints")
@@ -560,12 +560,12 @@ def show_anns(self, anns: List[Dict]):
     title="Color Map Processor",
     tags=["controlnet"],
     category="controlnet",
-    version="1.2.2",
+    version="1.2.3",
 )
 class ColorMapImageProcessorInvocation(ImageProcessorInvocation):
     """Generates a color map from the provided image"""
 
-    color_map_tile_size: int = InputField(default=64, ge=0, description=FieldDescriptions.tile_size)
+    color_map_tile_size: int = InputField(default=64, ge=1, description=FieldDescriptions.tile_size)
 
     def run_processor(self, image: Image.Image):
         np_image = np.array(image, dtype=np.uint8)
@@ -592,15 +592,15 @@ def run_processor(self, image: Image.Image):
     title="Depth Anything Processor",
     tags=["controlnet", "depth", "depth anything"],
     category="controlnet",
-    version="1.1.1",
+    version="1.1.2",
 )
 class DepthAnythingImageProcessorInvocation(ImageProcessorInvocation):
     """Generates a depth map based on the Depth Anything algorithm"""
 
     model_size: DEPTH_ANYTHING_MODEL_SIZES = InputField(
         default="small", description="The size of the depth model to use"
     )
-    resolution: int = InputField(default=512, ge=64, multiple_of=64, description=FieldDescriptions.image_res)
+    resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
     def run_processor(self, image: Image.Image):
         depth_anything_detector = DepthAnythingDetector()
@@ -615,15 +615,15 @@ def run_processor(self, image: Image.Image):
     title="DW Openpose Image Processor",
     tags=["controlnet", "dwpose", "openpose"],
     category="controlnet",
-    version="1.1.0",
+    version="1.1.1",
 )
 class DWOpenposeImageProcessorInvocation(ImageProcessorInvocation):
     """Generates an openpose pose from an image using DWPose"""
 
     draw_body: bool = InputField(default=True)
     draw_face: bool = InputField(default=False)
     draw_hands: bool = InputField(default=False)
-    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+    image_resolution: int = InputField(default=512, ge=1, description=FieldDescriptions.image_res)
 
     def run_processor(self, image: Image.Image):
         dw_openpose = DWOpenposeDetector()
@@ -642,15 +642,15 @@ def run_processor(self, image: Image.Image):
     title="Heuristic Resize",
     tags=["image, controlnet"],
     category="image",
-    version="1.0.0",
+    version="1.0.1",
     classification=Classification.Prototype,
 )
 class HeuristicResizeInvocation(BaseInvocation):
     """Resize an image using a heuristic method. Preserves edge maps."""
 
     image: ImageField = InputField(description="The image to resize")
-    width: int = InputField(default=512, gt=0, description="The width to resize to (px)")
-    height: int = InputField(default=512, gt=0, description="The height to resize to (px)")
+    width: int = InputField(default=512, ge=1, description="The width to resize to (px)")
+    height: int = InputField(default=512, ge=1, description="The height to resize to (px)")
 
     def invoke(self, context: InvocationContext) -> ImageOutput:
         image = context.images.get_pil(self.image.image_name, "RGB")

invokeai/app/invocations/latent.py

@@ -3,15 +3,14 @@
 import math
 from contextlib import ExitStack
 from functools import singledispatchmethod
-from typing import Any, Iterator, List, Literal, Optional, Tuple, Union
+from typing import Any, Dict, Iterator, List, Literal, Optional, Tuple, Union
 
 import einops
 import numpy as np
 import numpy.typing as npt
 import torch
 import torchvision
 import torchvision.transforms as T
-from diffusers import AutoencoderKL, AutoencoderTiny
 from diffusers.configuration_utils import ConfigMixin
 from diffusers.image_processor import VaeImageProcessor
 from diffusers.models.adapter import T2IAdapter
@@ -21,9 +20,12 @@
     LoRAXFormersAttnProcessor,
     XFormersAttnProcessor,
 )
+from diffusers.models.autoencoders.autoencoder_kl import AutoencoderKL
+from diffusers.models.autoencoders.autoencoder_tiny import AutoencoderTiny
 from diffusers.models.unets.unet_2d_condition import UNet2DConditionModel
-from diffusers.schedulers import DPMSolverSDEScheduler
-from diffusers.schedulers import SchedulerMixin as Scheduler
+from diffusers.schedulers.scheduling_dpmsolver_sde import DPMSolverSDEScheduler
+from diffusers.schedulers.scheduling_tcd import TCDScheduler
+from diffusers.schedulers.scheduling_utils import SchedulerMixin as Scheduler
 from PIL import Image, ImageFilter
 from pydantic import field_validator
 from torchvision.transforms.functional import resize as tv_resize
@@ -521,9 +523,10 @@ def _concat_regional_text_embeddings(
             )
 
         if is_sdxl:
-            return SDXLConditioningInfo(
-                embeds=text_embedding, pooled_embeds=pooled_embedding, add_time_ids=add_time_ids
-            ), regions
+            return (
+                SDXLConditioningInfo(embeds=text_embedding, pooled_embeds=pooled_embedding, add_time_ids=add_time_ids),
+                regions,
+            )
         return BasicConditioningInfo(embeds=text_embedding), regions
 
     def get_conditioning_data(
@@ -825,7 +828,7 @@ def init_scheduler(
         denoising_start: float,
         denoising_end: float,
         seed: int,
-    ) -> Tuple[int, List[int], int]:
+    ) -> Tuple[int, List[int], int, Dict[str, Any]]:
         assert isinstance(scheduler, ConfigMixin)
         if scheduler.config.get("cpu_only", False):
             scheduler.set_timesteps(steps, device="cpu")
@@ -853,13 +856,15 @@ def init_scheduler(
         timesteps = timesteps[t_start_idx : t_start_idx + t_end_idx]
         num_inference_steps = len(timesteps) // scheduler.order
 
-        scheduler_step_kwargs = {}
+        scheduler_step_kwargs: Dict[str, Any] = {}
         scheduler_step_signature = inspect.signature(scheduler.step)
         if "generator" in scheduler_step_signature.parameters:
             # At some point, someone decided that schedulers that accept a generator should use the original seed with
             # all bits flipped. I don't know the original rationale for this, but now we must keep it like this for
             # reproducibility.
-            scheduler_step_kwargs = {"generator": torch.Generator(device=device).manual_seed(seed ^ 0xFFFFFFFF)}
+            scheduler_step_kwargs.update({"generator": torch.Generator(device=device).manual_seed(seed ^ 0xFFFFFFFF)})
+        if isinstance(scheduler, TCDScheduler):
+            scheduler_step_kwargs.update({"eta": 1.0})
 
         return num_inference_steps, timesteps, init_timestep, scheduler_step_kwargs
 

invokeai/backend/stable_diffusion/schedulers/schedulers.py

@@ -13,6 +13,7 @@
     LCMScheduler,
     LMSDiscreteScheduler,
     PNDMScheduler,
+    TCDScheduler,
     UniPCMultistepScheduler,
 )
 
@@ -40,4 +41,5 @@
     "dpmpp_sde_k": (DPMSolverSDEScheduler, {"use_karras_sigmas": True, "noise_sampler_seed": 0}),
     "unipc": (UniPCMultistepScheduler, {"cpu_only": True}),
     "lcm": (LCMScheduler, {}),
+    "tcd": (TCDScheduler, {}),
 }