Update app.py

app.py

@@ -1,58 +1,26 @@
 import gradio as gr
 import pandas as pd
-from catboost import CatBoostClassifier
 import joblib
 
-# Load and prepare data
-train_df = pd.read_csv('/kaggle/input/amazon-employee-access-challenge/train.csv')
-X = train_df.drop('ACTION', axis=1)
-y = train_df['ACTION']
-
-# Train and save model
-def train_and_save_model():
-    model = CatBoostClassifier(
-        iterations=100,
-        learning_rate=0.1,
-        depth=6,
-        verbose=0,
-        task_type='CPU',
-        bootstrap_type='Bernoulli',
-        subsample=0.8,
-        eval_metric='Accuracy',
-        early_stopping_rounds=20
-    )
-    model.fit(X, y)
-    joblib.dump(model, 'amazon_access_model.joblib', compress=3)
-    return model
+# Load the saved model
+model = joblib.load('amazon_access_model.joblib')
 
-# Cache common values
-COMMON_VALUES = {
-    'ROLE_ROLLUP_1': train_df['ROLE_ROLLUP_1'].mode()[0],
-    'ROLE_ROLLUP_2': train_df['ROLE_ROLLUP_2'].mode()[0],
-    'ROLE_DEPTNAME': train_df['ROLE_DEPTNAME'].mode()[0],
-    'ROLE_FAMILY_DESC': train_df['ROLE_FAMILY_DESC'].mode()[0],
-    'ROLE_FAMILY': train_df['ROLE_FAMILY'].mode()[0],
-    'ROLE_CODE': train_df['ROLE_CODE'].mode()[0]
-}
-
-# Load or train model
-try:
-    model = joblib.load('amazon_access_model.joblib')
-except:
-    model = train_and_save_model()
+# Load minimal data just for dropdowns
+train_df = pd.read_csv('/kaggle/input/amazon-employee-access-challenge/train.csv')
 
 def predict_access(resource, mgr_id, role_title):
+    # Common values for other fields
     input_data = pd.DataFrame([[
         resource,
         mgr_id,
-        COMMON_VALUES['ROLE_ROLLUP_1'],
-        COMMON_VALUES['ROLE_ROLLUP_2'],
-        COMMON_VALUES['ROLE_DEPTNAME'],
+        train_df['ROLE_ROLLUP_1'].mode()[0],
+        train_df['ROLE_ROLLUP_2'].mode()[0],
+        train_df['ROLE_DEPTNAME'].mode()[0],
         role_title,
-        COMMON_VALUES['ROLE_FAMILY_DESC'],
-        COMMON_VALUES['ROLE_FAMILY'],
-        COMMON_VALUES['ROLE_CODE']
-    ]], columns=X.columns)
+        train_df['ROLE_FAMILY_DESC'].mode()[0],
+        train_df['ROLE_FAMILY'].mode()[0],
+        train_df['ROLE_CODE'].mode()[0]
+    ]], columns=train_df.columns[1:])  # Exclude ACTION column
     
     prediction = model.predict(input_data)[0]
     confidence = model.predict_proba(input_data)[0][prediction]
@@ -60,32 +28,17 @@ def predict_access(resource, mgr_id, role_title):
     result = "✅ Access Granted" if prediction == 1 else "❌ Access Denied"
     return f"{result} (Confidence: {confidence:.2%})"
 
-# Create Gradio interface
+# Simple interface
 iface = gr.Interface(
     fn=predict_access,
     inputs=[
-        gr.Dropdown(
-            choices=sorted(train_df['RESOURCE'].unique().tolist())[:100],  # Limit choices
-            label="Resource"
-        ),
-        gr.Dropdown(
-            choices=sorted(train_df['MGR_ID'].unique().tolist())[:100],  # Limit choices
-            label="Manager"
-        ),
-        gr.Dropdown(
-            choices=sorted(train_df['ROLE_TITLE'].unique().tolist()),
-            label="Role Title"
-        )
+        gr.Dropdown(choices=sorted(train_df['RESOURCE'].unique().tolist())[:100], label="Resource"),
+        gr.Dropdown(choices=sorted(train_df['MGR_ID'].unique().tolist())[:100], label="Manager"),
+        gr.Dropdown(choices=sorted(train_df['ROLE_TITLE'].unique().tolist()), label="Role Title")
     ],
     outputs=gr.Text(label="Access Decision"),
     title="Amazon Access Control",
-    description="Select employee details to check access permission",
-    theme="soft",
-    examples=[
-        [train_df['RESOURCE'].iloc[0], train_df['MGR_ID'].iloc[0], train_df['ROLE_TITLE'].iloc[0]],
-        [train_df['RESOURCE'].iloc[1], train_df['MGR_ID'].iloc[1], train_df['ROLE_TITLE'].iloc[1]]
-    ]
+    theme="soft"
 )
 
-if __name__ == "__main__":
-    iface.launch()
+iface.launch()