File size: 896 Bytes
379d8e8 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 | import streamlit as st
import pandas as pd
import numpy as np
from tensorflow.keras.models import load_model
from sklearn.preprocessing import StandardScaler
# Load the trained model
model = load_model('best_dnn_model')
# Load the scaler
scaler = StandardScaler()
# Define function to preprocess input data
def preprocess_data(data):
data = np.array(data).reshape(1, -1)
data = scaler.transform(data)
return data
# Streamlit app
st.title('Bank Churn: DNN Model Deployment')
# Collect user input
user_input = st.text_area("Enter your input data (comma-separated)")
# Process the input and make prediction
if st.button('Predict'):
try:
data = [float(i) for i in user_input.split(',')]
data = preprocess_data(data)
prediction = model.predict(data)
st.write(f"Prediction: {prediction}")
except Exception as e:
st.write(f"Error: {e}")
|