Create app.py

app.py

@@ -0,0 +1,30 @@
+import gradio as gr
+import numpy as np
+import onnxruntime as ort
+from PIL import Image
+
+# Load your ONNX model
+sess = ort.InferenceSession("visionguard_simplified.onnx", providers=["CPUExecutionProvider"])
+
+# Preprocess + inference
+def detect_corruption(img: Image.Image):
+    img = img.resize((128,128)).convert("RGB")
+    arr = np.array(img).astype(np.float32)/255.0
+    mean = np.array([0.485,0.456,0.406],dtype=np.float32)
+    std  = np.array([0.229,0.224,0.225],dtype=np.float32)
+    x = ((arr-mean)/std).transpose(2,0,1)[None,...]
+    logits = sess.run(None, {"input": x})[0]
+    prob   = float(1/(1+np.exp(-logits[0,0])))
+    return {"clean": 1-prob, "corrupted": prob}
+
+# Gradio interface
+iface = gr.Interface(
+    fn=detect_corruption,
+    inputs=gr.Image(type="pil"),
+    outputs=gr.Label(num_top_classes=2, label="Corruption Score"),
+    title="VisionGuard Corruption Detector",
+    description="Upload a frame, get corruption probabilities."
+)
+
+if __name__=="__main__":
+    iface.launch()