Divija89/Tips-predictor-model1

1import streamlit as st
2import pandas as pd
3import joblib
4import numpy as np
5
6st.set_page_config(page_title=" Taxi Tip Predictor", layout="centered")
7
8
9st.title("Tip prediction ")
10st.write("Enter the details below to predict the expected tip amount.")
11
12# Load the trained model (should be a pipeline)
13model = joblib.load("tips.pkl")
14
15# Streamlit UI to take inputs
16with st.form("tip_form"):
17    total_bill = st.slider("Total Bill ($)", min_value=0.0, max_value=500.00,value=20.0)
18    sex = st.selectbox("Sex", ["Male", "Female"])
19    smoker = st.selectbox("Smoker", ["Yes", "No"])
20    day = st.selectbox("Day", ["Thur", "Fri", "Sat", "Sun"])
21    time = st.selectbox("Time", ["Lunch", "Dinner"])
22    size = st.number_input("Party Size", min_value=1, value=2)
23
24    # Submit button
25    submitted = st.form_submit_button("Predict Tip")
26
27# Prediction on form submission
28if submitted:
29    input_df = pd.DataFrame([{
30        'total_bill': total_bill,
31        'sex': sex,
32        'smoker': smoker,
33        'day': day,
34        'time': time,
35        'size': size
36    }])
37
38    # Print input data
39    #st.write("Input Data:")
40    #st.dataframe(input_df)
41
42    # Check the model type again just before prediction
43    #st.write(f"Model type before prediction: {type(model)}")  # Should show <class 'sklearn.pipeline.Pipeline'>
44
45    try:
46        # Predict the tip
47        prediction = model.predict(input_df)
48
49        # Ensure the output is a scalar value
50        predicted_tip = prediction[0] if isinstance(prediction, (list, np.ndarray)) else prediction
51
52        # Display the predicted tip
53        st.success(f"Predicted Tip: *${predicted_tip:.2f}*")
54    except Exception as e:
55        
56        st.error(f" Error: {str(e)}")