[Python] How can I solve this on Streamlit: Error in input: The feature names should match...

PROJECT BY SAMUEL NNAMANI a.k.a Sammyst The Analyst.


The full error reads: Error in input: The feature names should match those that were passed during fit. Feature names unseen at fit time:

campaign day previous.

# Import the necessary dependencies
import streamlit as st
import pandas as pd
import plotly.express as px
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score`


# =============== PAGE CONFIGURATION ==============
st.set_page_config(page_title="Customer Segmentation & Loan Prediction Dashboard", layout="wide")

# =============== LOAD DATA =================
@st.cache_data
def load_data():
df = pd.read_csv("bank-full3.csv")
df.columns = df.columns.str.strip().str.lower().str.replace(" ", "_")
return df

# Load the data into a DataFrame
df = load_data()

# Show column names for debugging
# st.write("Available Columns:", df.columns.tolist())


# =============== PREPROCESSING ==============
# Convert target column 'y' to binary
df["y"] = df["y"].map({"yes":1, "no":0})

# Select relevant features for the model - mostly numeric features
features = ["age", "balance", "day", "duration", "campaign", "previous"]
X = df[features]
y = df["y"]

# Standardize the feature values
scaler = StandardScaler()
X_scaled_array = scaler.fit_transform(X)
X_scaled = pd.DataFrame(X_scaled_array, columns=features)

# Use the train_test_split function
X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.3, random_state=42)

# Train Random Forest model
rf_model = RandomForestClassifier()
rf_model.fit(X_train, y_train)

# Predict the outcome
# rf_pred = rf_model.predict(X_test)
# Print the accuracy_score
# st.write("The model score of this algorithm is:", accuracy_score(rf_pred, y_test))


# ================== SIDEBAR CONTROLS ==================
st.sidebar.header("Dashboard Controls")

features_selected = st.sidebar.multiselect(
"Features for Clustering:",
options=features,
default=['age', 'balance', 'duration']
)

n_clusters = st.sidebar.slider("Number of Segments (Clusters):", min_value=2, max_value=10, value=4)

# ====================== CLUSTERING =======================
X_cluster = df[features_selected]
X_cluster_scaled = scaler.fit_transform(X_cluster)

kmeans = KMeans(n_clusters=n_clusters, random_state=42)
clusters = kmeans.fit_predict(X_cluster_scaled)

df['customer_segment'] = clusters

segment_selected = st.sidebar.multiselect(
"Filter by Customer Segment:",
options=sorted(df["customer_segment"].unique()),
default=sorted(df["customer_segment"].unique())
)

df_selection = df[df["customer_segment"].isin(segment_selected)]


Here is the codes for what will be on the dashboard

# ================= MAIN DASHBOARD ======================
st.title("Customer Segmentation & Subscription Prediction Dashboard")

# KPIs
st.subheader("Key Performance Indicators (KPIs)")
kpi1, kpi2, kpi3 = st.columns(3)

with kpi1:
st.metric(label="Total Customers", value=df_selection.shape[0])

with kpi2:
st.metric(label="Average Balance", value=f"${df_selection['balance'].mean():.2f}")

with kpi3:
st.metric(label="Average Age", value=f"{df_selection['age'].mean():.1f} years")

st.markdown("---")

# Visualizations
st.subheader("Visualizations")

# Histogram of Customer segments
st.markdown("### Customer Segments Distribution")
fig_segment = px.histogram(df_selection, x="customer_segment", color="customer_segment")
st.plotly_chart(fig_segment, use_container_width=True)

# Subscription rate
st.markdown("### Subscription Rate by Segment")
fig_subscribe = px.histogram(df_selection, x='customer_segment', color='y', barmode='group')
st.plotly_chart(fig_subscribe, use_container_width=True)

# Job distribution
st.markdown("### Job Distribution across Segments")
fig_job = px.histogram(df_selection, x='job', color="customer_segment", barmode='group')
st.plotly_chart(fig_job, use_container_width=True)

# Pie Charts
st.subheader("Pie Charts")

pie_col1, pie_col2 = st.columns(2)

with pie_col1:
fig_pie_segment = px.pie(df_selection, names='customer_segment', title="Customer Segment Share")
st.plotly_chart(fig_pie_segment, use_container_width=True)

with pie_col2:
fig_pie_subscribe = px.pie(df_selection, names="y", title="Subscription (Yes/No) Share")
st.plotly_chart(fig_pie_subscribe, use_container_width=True)

st.markdown("---")

# Scatter Plot
st.subheader("Scatter Plot of Segments")

scatter_x = st.selectbox("Select X-axis:", features_selected, index=0)
scatter_y = st.selectbox("Select Y-axis:", features_selected, index=1)

fig_scatter = px.scatter(
df_selection,
x=scatter_x,
y=scatter_y,
color="customer_segment",
symbol="customer_segment",
title=f"{scatter_x} vs {scatter_y} by Segment"
)
st.plotly_chart(fig_scatter, use_container_width=True)


# ================== PREDICT CUSTOMER SUBSCRIPTION TO LOAN ====================
st.subheader("Predict Customer Subscription Probability")
st.write("This will predict the probability of a customer subscribing to a loan package by the financial institution")

with st.form('predict_form'):
age_input = st.number_input("Age:", min_value=18, max_value=100, value=30)
balance_input = st.number_input("Account Balance ($):", min_value=500, max_value=100000, value=500)
day_input = st.number_input("Day of Month Contacted:", min_value=1, max_value=31, value=15)
duration_input = st.number_input("Last Contact Duration (seconds):", min_value=1, max_value=5000, value=200)
campaign_input = st.number_input("Number of Contacts During Campaign:", min_value=1, max_value=50, value=3)
previous_input = st.number_input("Number of Previous Contacts:", min_value=0, max_value=50, value=0)

submit = st.form_submit_button("Predict")


I am suspecting that this is where my problem is coming from. But I cannot understand what is happening.

if submit:
try:
# SANITIZE all inputs: remove whitespace, then converts all of them to numeric
#age_input = int(str(age_input).strip())
#balance_input = float(str(balance_input).strip())
#day_input = int(str(day_input).strip())
#duration_input = int(str(duration_input).strip())
#campaign_input = int(str(campaign_input).strip())
#previous_input = int(str(previous_input).strip())

feature_names = ["age", "balance", "day", "duration", "campaign", "previous"]
user_data = pd.DataFrame([[age_input, balance_input, day_input, duration_input, campaign_input, previous_input]],
columns=feature_names)

# Create user input DataFrame with all required features
# user_data = pd.DataFrame({
# "age": [age_input],
# "balance": [balance_input],
# "day": [day_input],
# "duration": [duration_input],
# "campaign": [campaign_input],
# "previous": [previous_input]
# })

# Ensure the columns are in the same order as during training
# user_data = user_data[rf_model.feature_names_in_]
# Scale and ensure column alignment
user_data_scaled_array = scaler.transform(user_data)
user_data_scaled = pd.DataFrame(user_data_scaled_array, columns=feature_names)
user_data_scaled = user_data_scaled[rf_model.feature_names_in_]

# Predict
pred_prob = rf_model.predict_proba(user_data_scaled)[0][1] * 100
pred_class = "Subscribed" if pred_prob >= 50 else "Not Subscribed"

st.metric(label="Predicted Subscription Probability", value=f"{pred_prob:.2f}%")
st.success(f"Prediction: **{pred_class}**")

# Financial Health Score
st.subheader(" Financial Health Score")
if balance_input >= 5000:
score = " Excellent"
elif balance_input >= 2000:
score = " Good"
elif balance_input >= 0:
score = " Average"
else:
score = " Poor"

st.info(f"Financial Health: {score}")

except Exception as e:
st.error(f" Error in input: {e}. Please make sure all fields are numeric without strange characters.")
st.write("Model expects these features:", list(rf_model.feature_names_in_))

Continue reading...