From 2654fe78b9e5deb23f7e04b2996d66ce7d4334cf Mon Sep 17 00:00:00 2001
From: Pranav Kodlinge <pranavkodlinge@gmail.com>
Date: Wed, 2 Oct 2024 03:43:33 +0530
Subject: [PATCH] Updated and Improved Loan Estimator Using RF

I have updated the predict file in models\loan_eligibility and did esitmations using random forest model considering the real world scenario.
---
 models/loan_eligibility/model.py   |  6 ---
 models/loan_eligibility/predict.py | 79 ++++++++++++++++++++++++++++--
 2 files changed, 76 insertions(+), 9 deletions(-)
 delete mode 100644 models/loan_eligibility/model.py

diff --git a/models/loan_eligibility/model.py b/models/loan_eligibility/model.py
deleted file mode 100644
index 77695b88..00000000
--- a/models/loan_eligibility/model.py
+++ /dev/null
@@ -1,6 +0,0 @@
-def loan_eligibility(income, loan_amount, credit_score):
-	# Placeholder logic for loan eligibility prediction
-	if income > 50000 and credit_score[0] > 700:
-		return "Loan approved"
-	else:
-		return "Loan denied"
diff --git a/models/loan_eligibility/predict.py b/models/loan_eligibility/predict.py
index b593359e..79b09957 100644
--- a/models/loan_eligibility/predict.py
+++ b/models/loan_eligibility/predict.py
@@ -1,4 +1,77 @@
-from models.loan_eligibility.model import loan_eligibility
+import numpy as np
+import pandas as pd
+from sklearn.model_selection import train_test_split
+from sklearn.ensemble import RandomForestClassifier
+import streamlit as st
 
-def get_prediction(income, loan_amount, credit_score):
-	return loan_eligibility(income, loan_amount, credit_score)
\ No newline at end of file
+# Simulate some data (Loan Amount, Credit Score, Income)
+np.random.seed(42)
+data_size = 1000
+
+income = np.random.randint(0, 150001, size=data_size)  # Income between 0 and ₹1.5 Lakhs
+loan_amount = np.random.randint(5000, 50000, size=data_size)  # Loan amount between ₹5k and ₹50k
+credit_score = np.random.randint(300, 850, size=data_size)  # Credit score between 300 and 850
+
+# Derived feature: Loan-to-Income Ratio
+loan_to_income_ratio = loan_amount / (income + 1)  # Adding 1 to avoid division by zero
+
+# Function to determine eligibility 
+def is_eligible(income, loan_amount, credit_score):
+    # Simple rule-based eligibility logic
+    if income > 40000 and loan_amount < 0.3 * income and credit_score > 600:
+        return 1  # Eligible
+    else:
+        return 0  # Not Eligible
+
+eligibility = np.array([is_eligible(income[i], loan_amount[i], credit_score[i]) for i in range(data_size)])
+
+# Create a DataFrame for training
+df = pd.DataFrame({
+    'Income': income,
+    'Loan Amount': loan_amount,
+    'Credit Score': credit_score,
+    'Loan-to-Income Ratio': loan_to_income_ratio,
+    'Eligibility': eligibility
+})
+
+#Features and target
+X = df[['Income', 'Loan Amount', 'Credit Score', 'Loan-to-Income Ratio']]
+y = df['Eligibility']
+
+# Split the data into training and testing 
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
+
+#Training Random Forest model
+model = RandomForestClassifier(n_estimators=100, random_state=42)
+model.fit(X_train, y_train)
+
+# Step 4: Create the Streamlit app
+st.title("Loan Eligibility Predictor ")
+
+# User inputs
+income_input = st.number_input("Enter your Income (₹)", min_value=0, step=5000, format="%d")  # No upper cap on income
+
+# Calculate 80% of income as the maximum loan amount
+max_loan_amount = 0.8 * income_input if income_input > 0 else 5000
+
+# Ensure that loan amount is at least 5000 and dynamically adjust the upper limit
+loan_amount_input = st.number_input(f"Enter the Loan Amount you want (₹)", 
+                                    min_value=5000, 
+                                    max_value=int(max_loan_amount), 
+                                    step=1000)
+
+credit_score_input = st.number_input("Enter your Credit Score", min_value=300, max_value=850, step=10)
+
+# Derived feature: Loan-to-Income Ratio
+loan_to_income_ratio_input = loan_amount_input / (income_input + 1)  # Adding 1 to avoid division by zero
+
+# Prediction button
+if st.button("Check Eligibility"):
+    # Predict eligibility based on user inputs
+    new_data = np.array([[income_input, loan_amount_input, credit_score_input, loan_to_income_ratio_input]])
+    prediction = model.predict(new_data)
+    
+    if prediction == 1:
+        st.success("You are Eligible for the loan!")
+    else:
+        st.error("You are Not Eligible for the loan.")