Create app.py

app.py

@@ -0,0 +1,70 @@
+import gradio as gr
+import pickle
+import pandas as pd
+import joblib
+from tensorflow.keras.models import load_model
+
+
+# Load your pre-trained model
+model = load_model('/content/best_model.h5')
+# Define the prediction function
+def predict(age, job, marital, education, default, balance, housing, loan, contact, day, month, duration, campaign, pdays, previous, poutcome):
+    # Define the expected input order and preprocess accordingly
+    columns = [
+        'age', 'job', 'marital', 'education', 'default', 'balance', 'housing',
+        'loan', 'contact', 'day', 'month', 'duration', 'campaign', 'pdays',
+        'previous', 'poutcome'
+    ]
+
+    # Prepare the input values
+    data = [
+        age, job, marital, education, default, balance, housing, loan,
+        contact, day, month, duration, campaign, pdays, previous, poutcome
+    ]
+
+    # Convert to DataFrame
+    df = pd.DataFrame([data], columns=columns)
+
+    # Preprocess: One-hot encode categorical features (simulating as example)
+    # Normally, ensure you replicate the preprocessing steps used during training
+    df_processed = pd.get_dummies(df)
+
+    # Align processed DataFrame with model input (add missing columns if any)
+    model_columns = model.feature_names_in_  # Assuming the model has this attribute
+    for col in model_columns:
+        if col not in df_processed:
+            df_processed[col] = 0
+
+    df_processed = df_processed[model_columns]
+
+    # Predict
+    prediction = model.predict(df_processed)[0]
+
+    return "Yes" if prediction == 1 else "No"
+
+# Define Gradio interface
+inputs = [
+    gr.Number(label="Age"),
+    gr.Dropdown(['management', 'technician', 'entrepreneur', 'blue-collar', 'unknown', 'retired', 'admin.', 'services', 'self-employed', 'unemployed', 'student', 'housemaid'], label="Job"),
+    gr.Dropdown(['married', 'single', 'divorced'], label="Marital Status"),
+    gr.Dropdown(['primary', 'secondary', 'tertiary', 'unknown'], label="Education"),
+    gr.Dropdown(['yes', 'no'], label="Default"),
+    gr.Number(label="Balance"),
+    gr.Dropdown(['yes', 'no'], label="Housing Loan"),
+    gr.Dropdown(['yes', 'no'], label="Personal Loan"),
+    gr.Dropdown(['unknown', 'telephone', 'cellular'], label="Contact"),
+    gr.Number(label="Day"),
+    gr.Dropdown(['jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul', 'aug', 'sep', 'oct', 'nov', 'dec'], label="Month"),
+    gr.Number(label="Duration"),
+    gr.Number(label="Campaign"),
+    gr.Number(label="Pdays"),
+    gr.Number(label="Previous"),
+    gr.Dropdown(['unknown', 'other', 'failure', 'success'], label="Poutcome")
+]
+
+
+output = gr.Textbox(label="Subscription Prediction")
+
+gui = gr.Interface(fn=predict, inputs=inputs, outputs=output, title="Term Deposit Subscription Prediction")
+
+gui.launch()