Upload app.py

app.py

@@ -1,3 +1,4 @@
+
 import gradio as gr
 from joblib import load
 import pandas as pd
@@ -7,7 +8,6 @@ import pandas as pd
 dv , model = load("train_model.joblib")
 
 
-
 # creating a predict function to be passed into gradio UI
 def predict(age, job, marital, education, default, housing,
             loan, contact, month,day_of_week,campaign,pdays,
@@ -36,25 +36,62 @@ def predict(age, job, marital, education, default, housing,
     
     print(customer)
     df_transformed = dv.transform([customer])
-    prediction = model.predict_proba(df_transformed)[:,1]
+    prediction = model.predict_proba(df_transformed)[0]
     
+#     return prediction[1]
+    return {'Failure(!Deposit)': prediction[0], 'Success(Deposit)': prediction[1]}
+   
 #     Desposited = prediction >= 0.50
     
 #     result = {
 #         "deposit_probability": float(prediction),
 #         "Deposited": bool(Deposited)
 #     } 
-    print(f' The probabilty of depositing in the bank is  : {str(prediction)}')
+#     print(f' The probabilty of depositing in the bank is  : {str(prediction)}')
+    
     
-    return str(prediction)
 
 
+# In[4]:
+
+
+def feature_importance(age, job, marital, education, default, housing,
+            loan, contact, month,day_of_week,campaign,pdays,
+            previous,poutcome,emp_var_rate,cons_price_idx,cons_conf_idx):
+    customer = {
+        'age': age,
+        'job': job,
+        'marital': marital,
+        'education': education,
+        'default': default, 
+        'housing': housing,
+        'loan': loan,
+        'contact': contact,
+        'month': month,
+        'day_of_week': day_of_week,
+        'campaign': campaign,
+        'pdays': pdays,
+        'previous': previous,
+        'poutcome': poutcome,
+        'emp_var_rate': emp_var_rate,
+        'cons_price_idx': cons_price_idx,
+        'cons_conf_idx': cons_conf_idx,
+        
+    }
+    df_transformed = pd.DataFrame(dv.transform(customer))
+    df_transformed.columns = dv.get_feature_names_out()
+    important_features = pd.DataFrame({'cols':df_transformed.columns, 'imp':model.feature_importances_}
+                       ).sort_values('imp', ascending=False)
+    return important_features.plot('cols', 'imp', 'barh', figsize=(12,7), legend=False)
+#     return print(important_features.to_json())
+    
+
 
 
 age = gr.inputs.Slider(minimum=1,default = 35, maximum=100, step=1,label = 'Age') #default=data['age'].mean()
 job = gr.inputs.Dropdown(choices=["Housemaid", "Services","Admin.","Blue-Collar","Technician",
-                                 "Retired","Management","Unemployed","Self-Employed","Unknown",
-                                  "Entrepreneur","Student"],label = 'Job')
+                                 "Retired","Management","Unemployed","Self-Employed","Unknown","Entrepreneur","Student"],
+                         label = 'Job')
 marital = gr.inputs.Dropdown(choices=["Married", "Single","Divorced","Unknown"],label = 'Marital')
 education = gr.inputs.Dropdown(choices=["Basic.4y", "High.School","Basic.6y","Basic.9y","Professional.Course",
                                  "Unknown","University.Degree","Illiterate"],label = 'Education')
@@ -78,12 +115,7 @@ cons_conf_idx = gr.inputs.Slider(minimum=-51,default = -42, maximum=-27, step =
 iface = gr.Interface(predict,[age, job, marital, education, default, housing,
             loan, contact, month,day_of_week,campaign,pdays,
             previous,poutcome,emp_var_rate,cons_price_idx,cons_conf_idx],
-             outputs = "number",
-            interpretation="default",verbose = True
+            outputs = "label",
+                     interpretation=feature_importance
              )
-iface.launch(share=True)
-
-
-
-
-
+iface.launch(share=True , debug=True)