stage for update

.gitignore

@@ -0,0 +1,26 @@
+# Python cache files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# Jupyter Notebook checkpoints
+.ipynb_checkpoints
+
+# Environment variables
+.env
+
+# Streamlit specific files
+.streamlit/
+
+draft.ipynb
+
+# Model files
+
+*.h5
+
+# Logs
+*.log
+
+# Virtual environment
+venv/
+env/

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 keanteng
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

app.py

@@ -0,0 +1,103 @@
+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.")

backend/utils.py

@@ -0,0 +1,14 @@
+import google.generativeai as genai
+import pandas as pd
+
+def authenticate_gemini(api_key):
+    try:
+        genai.configure(api_key=api_key)
+        ai_model = genai.GenerativeModel("gemini-1.5-flash")
+        test = ai_model.generate_content("Explain how AI works")
+        return True
+    except Exception as e:
+        return False
+    
+def load_data(path):
+    return pd.read_csv(path)

git.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+
+# Get the current commit count
+commit_count=$(git rev-list --count HEAD)
+
+# Increment the commit count
+next_commit_count=$((commit_count + 1))
+
+# Add all changes
+git add .
+
+# Check if a custom message is provided
+if [ -z "$1" ]; then
+  commit_message="auto commit #$next_commit_count"
+else
+  commit_message="$1 #$next_commit_count"
+fi
+
+# Commit with the message
+git commit -m "$commit_message"
+
+# Push the changes
+git push
+
+# to run ./git_auto.sh "Your custom message"

model/model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07e0fdb36fa70e97c0bcaeb438056e7bb6d87f82af1df4c262ce1b1ada9c8ff4
+size 193308

pages/documentation.py

@@ -0,0 +1,20 @@
+import streamlit as st
+
+st.set_page_config(page_title='Documentation', layout='wide')
+
+with st.sidebar:
+    with st.expander("⚠️ Disclaimer", expanded=True):
+        st.write("This web app is intended for prediction purposes only. The results are based on the input data provided and \
+        the performance of the machine learning model. The accuracy of the predictions may vary depending on data quality \
+        and model reliability.")
+        
+    st.caption("MIT License 2025 © Khor Kean Teng, Loong Shih-Wai, Tioh Zi Cong, Yee See Marn")
+
+st.title("📄 Documentation")
+st.markdown("""
+            To learn more about the project, please refer to the sections below.
+            """)
+st.subheader("About Telco Churn")
+st.write("""Customer churn is a critical issue in the telecommunications industry. \
+    It refers to the percentage of customers who discontinue their services with a company within a given time period. 
+    The churn rate is a key metric for businesses to measure customer satisfaction and loyalty.""")

requirements.txt

@@ -0,0 +1,6 @@
+streamlit
+pandas
+joblib
+scikit-learn==1.3.1
+google-generativeai
+matplotlib

run.sh

@@ -0,0 +1,2 @@
+# only work if use python 3.12 on bash shell
+py -3.12 -m streamlit run app.py

test.py

@@ -0,0 +1,17 @@
+import joblib
+import pandas as pd
+
+# Load the model
+model = joblib.load('model/model.pkl')
+
+# Load the data
+data = pd.read_csv('data/sample_data.csv')
+
+# Make predictions
+predictions = model.predict(data)
+
+# save the predictions
+data['Churn'] = predictions
+
+# save as sample_output.csv
+data.to_csv('data/sample_output.csv', index=False)