feat(diffusers): update, add autopipeline, controlnet (mudler#1432)

* feat(diffusers): update, add autopipeline, controlenet

* tests with AutoPipeline

* simplify logic

Makefile

@@ -383,7 +383,7 @@ help: ## Show this help.
 protogen: protogen-go protogen-python
 
 protogen-go:
-	protoc --go_out=. --go_opt=paths=source_relative --go-grpc_out=. --go-grpc_opt=paths=source_relative \
+	protoc -Ibackend/ --go_out=pkg/grpc/proto/ --go_opt=paths=source_relative --go-grpc_out=pkg/grpc/proto/ --go-grpc_opt=paths=source_relative \
     backend/backend.proto
 
 protogen-python:

api/backend/image.go

@@ -27,6 +27,7 @@ func ImageGeneration(height, width, mode, step, seed int, positive_prompt, negat
 			CLIPModel:     c.Diffusers.ClipModel,
 			CLIPSubfolder: c.Diffusers.ClipSubFolder,
 			CLIPSkip:      int32(c.Diffusers.ClipSkip),
+			ControlNet:    c.Diffusers.ControlNet,
 		}),
 	})
 

api/config/config.go

@@ -38,8 +38,7 @@ type Config struct {
 
 	// Diffusers
 	Diffusers Diffusers `yaml:"diffusers"`
-
-	Step int `yaml:"step"`
+	Step      int       `yaml:"step"`
 
 	// GRPC Options
 	GRPC GRPC `yaml:"grpc"`
@@ -77,6 +76,7 @@ type Diffusers struct {
 	ClipSkip         int     `yaml:"clip_skip"`         // Skip every N frames
 	ClipModel        string  `yaml:"clip_model"`        // Clip model to use
 	ClipSubFolder    string  `yaml:"clip_subfolder"`    // Subfolder to use for clip model
+	ControlNet       string  `yaml:"control_net"`
 }
 
 type LLMConfig struct {

backend/backend.proto

@@ -110,6 +110,7 @@ message ModelOptions {
   string CLIPModel = 31;
   string CLIPSubfolder = 32;
   int32 CLIPSkip = 33;
+  string ControlNet = 48;
 
   // RWKV
   string Tokenizer = 34;

backend/python/diffusers/backend_diffusers.py

@@ -18,9 +18,9 @@
 import grpc
 
 from diffusers import StableDiffusionXLPipeline, StableDiffusionDepth2ImgPipeline, DPMSolverMultistepScheduler, StableDiffusionPipeline, DiffusionPipeline, EulerAncestralDiscreteScheduler
-from diffusers import StableDiffusionImg2ImgPipeline
+from diffusers import StableDiffusionImg2ImgPipeline, AutoPipelineForText2Image, ControlNetModel
 from diffusers.pipelines.stable_diffusion import safety_checker
-
+from diffusers.utils import load_image
 from compel import Compel
 
 from transformers import CLIPTextModel
@@ -30,6 +30,7 @@
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 COMPEL=os.environ.get("COMPEL", "1") == "1"
 CLIPSKIP=os.environ.get("CLIPSKIP", "1") == "1"
+SAFETENSORS=os.environ.get("SAFETENSORS", "1") == "1"
 
 # If MAX_WORKERS are specified in the environment use it, otherwise default to 1
 MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
@@ -135,8 +136,11 @@ def LoadModel(self, request, context):
             print(f"Loading model {request.Model}...", file=sys.stderr)
             print(f"Request {request}", file=sys.stderr)
             torchType = torch.float32
+            variant = None
+
             if request.F16Memory:
                 torchType = torch.float16
+                variant="fp16"
 
             local = False
             modelFile = request.Model
@@ -160,14 +164,8 @@ def LoadModel(self, request, context):
 
             fromSingleFile = request.Model.startswith("http") or request.Model.startswith("/") or local
 
-            if request.IMG2IMG and request.PipelineType == "":
-                request.PipelineType == "StableDiffusionImg2ImgPipeline"
-
-            if request.PipelineType == "":
-                request.PipelineType == "StableDiffusionPipeline"
-
             ## img2img
-            if request.PipelineType == "StableDiffusionImg2ImgPipeline":
+            if (request.PipelineType == "StableDiffusionImg2ImgPipeline") or (request.IMG2IMG and request.PipelineType == ""):
                 if fromSingleFile:
                     self.pipe = StableDiffusionImg2ImgPipeline.from_single_file(modelFile,
                                 torch_dtype=torchType,
@@ -177,12 +175,18 @@ def LoadModel(self, request, context):
                                 torch_dtype=torchType,
                                 guidance_scale=cfg_scale)
 
-            if request.PipelineType == "StableDiffusionDepth2ImgPipeline":
+            elif request.PipelineType == "StableDiffusionDepth2ImgPipeline":
                 self.pipe = StableDiffusionDepth2ImgPipeline.from_pretrained(request.Model,
                             torch_dtype=torchType,
                             guidance_scale=cfg_scale)
             ## text2img
-            if request.PipelineType == "StableDiffusionPipeline":
+            elif request.PipelineType == "AutoPipelineForText2Image" or request.PipelineType == "":
+                self.pipe = AutoPipelineForText2Image.from_pretrained(request.Model,
+                                                    torch_dtype=torchType,
+                                                    use_safetensors=SAFETENSORS, 
+                                                    variant=variant,
+                                                    guidance_scale=cfg_scale)
+            elif request.PipelineType == "StableDiffusionPipeline":
                 if fromSingleFile:
                     self.pipe = StableDiffusionPipeline.from_single_file(modelFile,
                                                         torch_dtype=torchType,
@@ -191,13 +195,11 @@ def LoadModel(self, request, context):
                     self.pipe = StableDiffusionPipeline.from_pretrained(request.Model,
                                                         torch_dtype=torchType,
                                                         guidance_scale=cfg_scale)
-
-            if request.PipelineType == "DiffusionPipeline":
+            elif request.PipelineType == "DiffusionPipeline":
                 self.pipe = DiffusionPipeline.from_pretrained(request.Model,
                                                         torch_dtype=torchType,
                                                         guidance_scale=cfg_scale)
-
-            if request.PipelineType == "StableDiffusionXLPipeline":
+            elif request.PipelineType == "StableDiffusionXLPipeline":
                 if fromSingleFile:
                     self.pipe = StableDiffusionXLPipeline.from_single_file(modelFile,
                                                                torch_dtype=torchType, use_safetensors=True,
@@ -207,21 +209,34 @@ def LoadModel(self, request, context):
                         request.Model, 
                         torch_dtype=torchType, 
                         use_safetensors=True, 
-                #       variant="fp16"
+                        variant=variant,
                         guidance_scale=cfg_scale)
-            # https://github.com/huggingface/diffusers/issues/4446
-            # do not use text_encoder in the constructor since then
-            # https://github.com/huggingface/diffusers/issues/3212#issuecomment-1521841481
+
             if CLIPSKIP and request.CLIPSkip != 0:
-                text_encoder = CLIPTextModel.from_pretrained(clipmodel, num_hidden_layers=request.CLIPSkip,  subfolder=clipsubfolder, torch_dtype=torchType)
-                self.pipe.text_encoder=text_encoder
+                self.clip_skip = request.CLIPSkip
+            else:
+                self.clip_skip = 0
+
             # torch_dtype needs to be customized. float16 for GPU, float32 for CPU
             # TODO: this needs to be customized
             if request.SchedulerType != "":
                 self.pipe.scheduler = get_scheduler(request.SchedulerType, self.pipe.scheduler.config)
+
             self.compel = Compel(tokenizer=self.pipe.tokenizer, text_encoder=self.pipe.text_encoder)
+
+
+            if request.ControlNet:
+                self.controlnet = ControlNetModel.from_pretrained(
+                    request.ControlNet, torch_dtype=torchType, variant=variant
+                )
+                self.pipe.controlnet = self.controlnet
+            else:
+                self.controlnet = None
+
             if request.CUDA:
                 self.pipe.to('cuda')
+                if self.controlnet:
+                    self.controlnet.to('cuda')
             # Assume directory from request.ModelFile.
             # Only if request.LoraAdapter it's not an absolute path
             if request.LoraAdapter and request.ModelFile != "" and not os.path.isabs(request.LoraAdapter) and request.LoraAdapter:
@@ -316,9 +331,15 @@ def GenerateImage(self, request, context):
             "num_inference_steps": steps,
         }
 
-        if request.src != "":
+        if request.src != "" and not self.controlnet:
             image = Image.open(request.src)
             options["image"] = image
+        elif self.controlnet and request.src:
+            pose_image = load_image(request.src)
+            options["image"] = pose_image
+
+        if CLIPSKIP and self.clip_skip != 0:
+            options["clip_skip"]=self.clip_skip
 
         # Get the keys that we will build the args for our pipe for
         keys = options.keys()