Update app.py

app.py

@@ -4,6 +4,18 @@ import joblib
 # Load the trained model
 model = joblib.load('trained_random_forest_pipeline.joblib')
 
+def predict(age, sex, race, etiology, hepatorenal_syndrome, omeprazole, spironolactone, furosemide, propanolol, dialysis, portal_vein_thrombosis,
+            ascites, hepatocellular_carcinoma, albumin, total_bilirubin, direct_bilirubin, inr, creatinine, platelets, ast, alt, hemoglobin, 
+            hematocrit, leukocytes, sodium, potassium, varices, red_wale_marks, rupture_point, active_bleeding, therapy, terlipressin_dose, rebleeding):
+    # Convert input data to the format expected by the model, e.g., a list or a DataFrame
+    input_data = [age, sex, race, etiology, hepatorenal_syndrome, omeprazole, spironolactone, furosemide, propanolol, dialysis, portal_vein_thrombosis,
+                  ascites, hepatocellular_carcinoma, albumin, total_bilirubin, direct_bilirubin, inr, creatinine, platelets, ast, alt, hemoglobin,
+                  hematocrit, leukocytes, sodium, potassium, varices, red_wale_marks, rupture_point, active_bleeding, therapy, terlipressin_dose, rebleeding]
+    # Assuming the model expects a single sample reshaped as a 2D array
+    prediction = model.predict([input_data])
+    return str(prediction[0])  # Convert prediction to string if necessary
+
+
 iface = gr.Interface(
     fn=predict,
     inputs=[
@@ -44,24 +56,5 @@ iface = gr.Interface(
     outputs="text"
 )
 
-
-
-
-def predict(age, sex, race, etiology, hepatorenal_syndrome, omeprazole, spironolactone, furosemide, propanolol, dialysis, portal_vein_thrombosis,
-            ascites, hepatocellular_carcinoma, albumin, total_bilirubin, direct_bilirubin, inr, creatinine, platelets, ast, alt, hemoglobin, 
-            hematocrit, leukocytes, sodium, potassium, varices, red_wale_marks, rupture_point, active_bleeding, therapy, terlipressin_dose, rebleeding):
-    # Convert input data to the format expected by the model, e.g., a list or a DataFrame
-    input_data = [age, sex, race, etiology, hepatorenal_syndrome, omeprazole, spironolactone, furosemide, propanolol, dialysis, portal_vein_thrombosis,
-                  ascites, hepatocellular_carcinoma, albumin, total_bilirubin, direct_bilirubin, inr, creatinine, platelets, ast, alt, hemoglobin,
-                  hematocrit, leukocytes, sodium, potassium, varices, red_wale_marks, rupture_point, active_bleeding, therapy, terlipressin_dose, rebleeding]
-    # Assuming the model expects a single sample reshaped as a 2D array
-    prediction = model.predict([input_data])
-    return str(prediction[0])  # Convert prediction to string if necessary
-
-
-
-
-
-
 if __name__ == "__main__":
     iface.launch()