Create app.py

app.py

@@ -0,0 +1,59 @@
+import gradio as gr
+from gradio_client import Client, handle_file
+import json
+import tempfile
+
+# Initialize both external Spaces
+client1 = Client("raqiat123/crop_disease_detection")
+client2 = Client("SoraRyuu/cv_first")
+
+def combined_predict(image):
+    # Temporarily save uploaded file
+    with tempfile.NamedTemporaryFile(suffix=".png", delete=False) as tmp:
+        image.save(tmp.name)
+        tmp_path = tmp.name
+
+    # Run prediction on both models
+    result1 = client1.predict(image=handle_file(tmp_path), api_name="/predict")
+    result2 = client2.predict(image=handle_file(tmp_path), api_name="/predict")
+
+    # Extract confidence values (adjust if needed)
+    conf1 = result1.get("confidence") if isinstance(result1, dict) else 0
+    conf2 = result2.get("confidence") if isinstance(result2, dict) else 0
+
+    if conf1 >= conf2:
+        chosen = {
+            "chosen_model": "raqiat123/crop_disease_detection",
+            "confidence": conf1,
+            "output": result1
+        }
+        result_text = f"Model Selected: crop_disease_detection\nConfidence: {conf1}"
+    else:
+        chosen = {
+            "chosen_model": "SoraRyuu/cv_first",
+            "confidence": conf2,
+            "output": result2
+        }
+        result_text = f"Model Selected: cv_first\nConfidence: {conf2}"
+
+    return result_text, json.dumps(chosen, indent=4)
+
+
+# Build Gradio UI
+with gr.Blocks() as demo:
+    gr.Markdown("# 🌱 Crop Classifier")
+    gr.Markdown("Upload an image. The Space will return the disease detected.")
+
+    image_input = gr.Image(type="pil")
+    text_output = gr.Textbox(label="Result (Text)")
+    json_output = gr.JSON(label="Raw JSON Result")
+
+    btn = gr.Button("Run Prediction")
+
+    btn.click(
+        fn=combined_predict,
+        inputs=image_input,
+        outputs=[text_output, json_output]
+    )
+
+demo.launch()