te

handler.py

@@ -8,21 +8,17 @@ from diffusers.utils import load_image
 
 import torch
 import torch.nn.functional as F
-from torchvision.transforms import Compose
+# from torchvision.transforms import Compose
 
 import PIL
 from PIL import Image
 
-from depth_anything.dpt import DepthAnything
-from depth_anything.util.transform import Resize, NormalizeImage, PrepareForNet
+# from depth_anything.dpt import DepthAnything
+# from depth_anything.util.transform import Resize, NormalizeImage, PrepareForNet
 
 from insightface.app import FaceAnalysis
 from pipeline_stable_diffusion_xl_instantid_full import StableDiffusionXLInstantIDPipeline, draw_kps
 from controlnet_aux import OpenposeDetector
-from huggingface_hub import hf_hub_download
-
-
-# end_test
 
 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 if device.type != 'cuda':
@@ -32,15 +28,8 @@ dtype = torch.float16 if str(device).__contains__("cuda") else torch.float32
 
 class EndpointHandler():
     def __init__(self, model_dir):
-        # hf_hub_download(repo_id="InstantX/InstantID", filename="ControlNetModel/config.json", local_dir="./checkpoints")
-        # hf_hub_download(
-        #     repo_id="InstantX/InstantID",
-        #     filename="ControlNetModel/diffusion_pytorch_model.safetensors",
-        #     local_dir="./checkpoints",
-        # )
-        # hf_hub_download(repo_id="InstantX/InstantID", filename="ip-adapter.bin", local_dir="./checkpoints")
-
-        print("Model dir: ", model_dir)
+        print("Loading model from", model_dir)
+        
         face_adapter = f"./checkpoints/ip-adapter.bin"
         controlnet_path = f"./checkpoints/ControlNetModel"
 
@@ -149,23 +138,23 @@ class EndpointHandler():
         self.generator = torch.Generator(device=device.type).manual_seed(3)
 
 
-        identitynet_strength_ratio = 0.8
-        pose_strength = 0.4
+        # identitynet_strength_ratio = 0.8
+        # pose_strength = 0.4
         self.my_controlnet_selection = ["pose"]
 
 
-        controlnet_scales = {
-            "pose": pose_strength,
-            # "canny": canny_strength,
-            # "depth": depth_strength,
-        }
-        self.pipe.controlnet = MultiControlNetModel(
-            [self.controlnet_identitynet]
-            + [self.controlnet_map[s] for s in self.my_controlnet_selection]
-        )
-        self.control_scales = [float(identitynet_strength_ratio)] + [
-            controlnet_scales[s] for s in self.my_controlnet_selection
-        ]
+        # controlnet_scales = {
+        #     "pose": pose_strength,
+        #     # "canny": canny_strength,
+        #     # "depth": depth_strength,
+        # }
+        # self.pipe.controlnet = MultiControlNetModel(
+        #     [self.controlnet_identitynet]
+        #     + [self.controlnet_map[s] for s in self.my_controlnet_selection]
+        # )
+        # self.control_scales = [float(identitynet_strength_ratio)] + [
+        #     controlnet_scales[s] for s in self.my_controlnet_selection
+        # ]
 
     def __call__(self, data):