diff --git a/README.md b/README.md
new file mode 100644
index 0000000..432d7df
--- /dev/null
+++ b/README.md
@@ -0,0 +1,28 @@
+# ComfyUI OOTDiffusion
+
+A ComfyUI custom node that simply integrates the OOTDiffusion functionality.
+
+一个简单接入 OOTDiffusion 的 ComfyUI 节点。
+
+## Instruction 指南
+
+根据 https://git-lfs.com 安装 git lfs：
+
+```txt
+sudo apt install git-lfs
+git lfs install
+```
+
+拉取 Huggingface 库至 `models/OOTDiffusion` 目录：
+
+```txt
+git clone https://huggingface.co/levihsu/OOTDiffusion models/OOTDiffusion
+```
+
+启动 ComfyUI 即可。
+
+## 节点
+
+LoadOOTDPipeline: 加载 OOTD Pipeline
+
+OOTDGenerate: 生成图像
diff --git a/humanparsing/parsing_api.py b/humanparsing/parsing_api.py
index 9ddecff..0eeccac 100644
--- a/humanparsing/parsing_api.py
+++ b/humanparsing/parsing_api.py
@@ -124,7 +124,8 @@ def load_atr_model():
     label = ['Background', 'Hat', 'Hair', 'Sunglasses', 'Upper-clothes', 'Skirt', 'Pants', 'Dress', 'Belt',
              'Left-shoe', 'Right-shoe', 'Face', 'Left-leg', 'Right-leg', 'Left-arm', 'Right-arm', 'Bag', 'Scarf']
     model = networks.init_model('resnet101', num_classes=num_classes, pretrained=None)
-    state_dict = torch.load('models/humanparsing/exp-schp-201908301523-atr.pth')['state_dict']
+    #! annotated config
+    state_dict = torch.load('models/OOTDiffusion/checkpoints/humanparsing/exp-schp-201908301523-atr.pth')['state_dict']
     new_state_dict = OrderedDict()
     for k, v in state_dict.items():
         name = k[7:]  # remove `module.`
@@ -142,7 +143,8 @@ def load_lip_model():
                   'Socks', 'Pants', 'Jumpsuits', 'Scarf', 'Skirt', 'Face', 'Left-arm', 'Right-arm',
                   'Left-leg', 'Right-leg', 'Left-shoe', 'Right-shoe']
     model = networks.init_model('resnet101', num_classes=num_classes, pretrained=None)
-    state_dict = torch.load('models/humanparsing/exp-schp-201908261155-lip.pth')['state_dict']
+    #! annotated config
+    state_dict = torch.load('models/OOTDiffusion/checkpoints/humanparsing/exp-schp-201908261155-lip.pth')['state_dict']
     new_state_dict = OrderedDict()
     for k, v in state_dict.items():
         name = k[7:]  # remove `module.`