File size: 4,623 Bytes
a17bdfd |
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 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 |
import streamlit as st
from backend.utils import *
import pandas as pd
from datetime import datetime, timedelta
import time
import joblib
import google.generativeai as genai
import matplotlib.pyplot as plt
# page layout
st.set_page_config(page_title="Telco Churn Engine", page_icon="🧊", layout="wide")
# title
st.title("Telco Churn Engine")
st.write("Speed up predicting customer churn in the telecommunications industry. Powered by Generative AI and Job Schedule Function.")
# sidebar
with st.sidebar:
with st.expander("⏰ Schedule Run (Demo)", expanded=False):
st.caption("Schedule a run for the app.")
run_date = st.date_input("Select Date")
run_time = st.time_input("Select Time")
countdown_placeholder = st.empty()
if st.button("Schedule Run", type='secondary'):
run_datetime = datetime.combine(run_date, run_time)
# scheduler.add_job(scheduled_task, 'date', run_date=run_datetime)
st.success(f"App scheduled to run on {run_datetime}.")
# Countdown logic
while True:
now = datetime.now()
time_left = run_datetime - now
if time_left.total_seconds() <= 0:
countdown_placeholder.write("Scheduled task is running!")
break
countdown_placeholder.write(f"Time left: {time_left}")
time.sleep(1)
with st.expander("⚙️ Generative AI", expanded=True):
st.caption("API token can be obtained at https://aistudio.google.com/.")
gemini_api = st.text_input("Gemini Token", "", type='password')
if authenticate_gemini(gemini_api):
st.success("Gemini API token is valid.")
else:
st.error("Gemini API token is invalid.")
with st.expander("🗳️ Sample Data Download", expanded=False):
st.caption("Download sample data for testing.")
sample_data = load_data("data/sample_data.csv")
st.download_button("Download Sample Data", sample_data.to_csv(), "sample_data.csv", "text/csv")
st.divider()
st.caption("MIT License 2025 © Khor Kean Teng, Loong Shih-Wai, Tioh Zi Cong, Yee See Marn")
# main content
with st.chat_message("assistant", avatar="https://static.vecteezy.com/system/resources/previews/035/010/451/non_2x/bionic-zombie-infusion-design-zombie-cyborg-evolution-icon-vector.jpg"):
response = st.write("Hello admin! I am Arnold. How can I automate so that you might lost your job?")
st.caption("If you use predefined data, the file upload step will be hidden.")
toggle = st.toggle('Use Predefined Data', True)
data = load_data("data/sample_data.csv")
if toggle == False:
uploaded_file = st.file_uploader("Upload a CSV file", type=["csv"])
if uploaded_file is not None:
data = pd.read_csv(uploaded_file)
submit = st.button("Execute", type='primary')
if submit:
# show preview in table in expander
with st.status("Preview Data", expanded=True):
st.write(data.head())
model = joblib.load("model/model.pkl")
prediction = model.predict(data)
data["Churn Prediction"] = prediction
# count how many churn
churn_count = data["Churn Prediction"].value_counts()
# show prview in table in expander
with st.status("Prediction", expanded=True):
st.write("The prediction is done. There are {} churn customers out of the total {} customers.".format(churn_count[1], len(data)))
st.write(data.head())
# plot a pie chart
with st.status("Churn Pie Chart", expanded=True):
st.write("The pie chart shows the distribution of churn customers.")
fig, ax = plt.subplots()
# resize the pie chart
fig.set_size_inches(3, 3)
ax.pie(churn_count, labels=["Churn", "Non-Churn"], autopct='%1.1f%%', startangle=90)
st.pyplot(fig)
with st.status("AI Opinion", expanded=True):
try:
ai_model = genai.GenerativeModel("gemini-1.5-flash")
response = ai_model.generate_content(f"Give some opinions in about 100 word based on the prediction results where there are {churn_count[1]} cases of attrition out of the total {len(data)} number of customers.")
st.write(response.text)
except Exception as e:
st.write("You don't have access to this feature. Please authenticate to use this feature.") |