Add config.json and model card to enable Inference API

README.md

@@ -0,0 +1,11 @@
+
+---
+license: mit
+pipeline_tag: image-segmentation
+---
+
+# DANet for Fracture Segmentation
+
+This is a DANet model with a ResNet-50 backbone, fine-tuned for segmenting bone fractures in X-ray images.
+
+This model was trained in a Kaggle notebook as part of a hackathon project.

config.json

@@ -0,0 +1,9 @@
+{
+  "architectures": [
+    "DANet"
+  ],
+  "auto_map": {
+    "AutoModel": "model.DANet"
+  },
+  "model_type": "danet"
+}

model.py

@@ -1,4 +1,5 @@
 
+# Paste the full DANet architecture code here... (same as before)
 import torch
 import torch.nn as nn
 from torch.nn import functional as F