Upload 4 files

dataset.csv

@@ -0,0 +1,44 @@
+Timestamp,1. Name ,2. How interested are you in the event topic?  ,  3. How close are you to the event location?  ,4. How many similar events have you attended in the past year? ,  5. How much free time do you have during the event timing?  ,  6. Are you willing to attend this event?  
+6/21/2025 9:52:30,Luffy,2,1,5,2,Yes
+6/21/2025 9:53:07,S.Shalini,2,2,1,1,Yes
+6/21/2025 9:53:12,Anachi,2,1,100,2,Yes
+6/21/2025 9:53:57,Madhumitha,2,1,2,2,Yes
+6/21/2025 9:54:16,Anantha Krishnan R ,2,1,2,2,Yes
+6/21/2025 9:54:27,Stark,1,2,19,2,Prefer not to Say
+6/21/2025 9:54:55,Reehan,2,1,50,2,Yes
+6/21/2025 9:55:58,Ram,1,2,2,2,Yes
+6/21/2025 9:57:24,Deepika ,2,2,2,2,No
+6/21/2025 9:59:29,Hari,2,2,59,2,No
+6/21/2025 10:00:40,prabhu,2,1,3,2,No
+6/21/2025 10:00:44,Rajesh,2,1,45,1,Prefer not to Say
+6/21/2025 10:01:00,sonia,2,2,3,2,Yes
+6/21/2025 10:01:14,zoro,1,2,78,2,Yes
+6/21/2025 10:01:39,Sanji,1,2,72,2,No
+6/21/2025 10:01:40,pradeep,1,1,4,2,No
+6/21/2025 10:01:44,yoga chandra shir,2,2,4,2,Yes
+6/21/2025 10:02:04,Shanks,2,1,8,2,Yes
+6/21/2025 10:02:12,vairaselvi,1,1,2,1,No
+6/21/2025 10:02:28,Manikandan.K,2,2,7,2,Yes
+6/21/2025 10:02:29,Aswin,1,1,3,1,Yes
+6/21/2025 10:02:39,nami,1,2,45,2,No
+6/21/2025 10:03:13,pavithra.V,2,1,7,1,No
+6/21/2025 10:04:00,charan.s,2,1,6,2,Yes
+6/21/2025 10:04:24,vishali,1,1,3,1,Yes
+6/21/2025 10:04:28,Suriya,2,2,10,2,Yes
+6/21/2025 10:04:45,surya prakash.s,2,1,9,2,No
+6/21/2025 10:05:06,Naruto,1,1,5,1,Yes
+6/21/2025 10:05:24,saranaya,2,2,5,2,No
+6/21/2025 10:05:34,Stark,2,1,60,2,Prefer not to Say
+6/21/2025 10:05:57,Garp,1,2,16,2,No
+6/21/2025 10:06:03,mahesh,2,1,5,2,Yes
+6/21/2025 10:06:19,Vivi,2,1,30,2,Yes
+6/21/2025 10:06:32,malathi,2,1,6,1,Yes
+6/21/2025 10:07:04,durga,2,2,6,1,Yes
+6/21/2025 10:07:14,PraveenD,2,2,50,2,Yes
+6/21/2025 10:07:38,Guts,1,2,50,2,Yes
+6/21/2025 10:07:40,pandi selvi,2,2,5,2,Yes
+6/21/2025 10:08:17,Ananth,1,2,20,2,Prefer not to Say
+6/21/2025 10:08:26,kaviya,1,1,8,2,Yes
+6/21/2025 10:08:54,Sakthi,2,2,89,2,Yes
+6/21/2025 10:09:10,parvesh,2,1,7,2,Yes
+6/21/2025 10:09:13,Sisa,2,1,30,2,Yes

ml_code.py

@@ -0,0 +1,170 @@
+# -*- coding: utf-8 -*-
+"""ml_code.ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1tETflt1JmWJudI-SDbtBiVIqjgCnaSU8
+
+Import Packages :
+"""
+
+import pandas as pd
+import numpy as np
+from sklearn.linear_model import LogisticRegression
+from sklearn.semi_supervised import SelfTrainingClassifier
+from sklearn.preprocessing import StandardScaler
+import joblib
+
+"""Dataset Load :"""
+
+# Load dataset
+df = pd.read_csv("data.csv")
+
+# Clean column names
+df.columns = df.columns.str.strip()
+
+# Rename relevant columns
+df = df.rename(columns={
+    '2. How interested are you in the event topic?': 'interest',
+    '3. How close are you to the event location?': 'proximity',
+    '4. How many similar events have you attended in the past year?': 'past_attendance',
+    '5. How much free time do you have during the event timing?': 'free_time',
+    '6. Are you willing to attend this event?': 'willingness'
+})
+
+df.head()
+
+"""Dataset Preprocess :"""
+
+# Normalize values from linear scale 1–2 → 0–1
+df[['interest', 'proximity', 'free_time']] = df[['interest', 'proximity', 'free_time']].apply(lambda x: (x - 1) / (2 - 1))
+
+# Convert willingness to numerical values
+df['willingness'] = df['willingness'].map({'Yes': 1, 'No': 0})
+
+# Features and labels
+X = df[['interest', 'proximity', 'past_attendance', 'free_time']].values
+y = df['willingness'].values
+
+"""Dataset into unlabeled :"""
+
+# Unlabeled samples = -1
+y_semi = np.array([label if not np.isnan(label) else -1 for label in y])
+
+# Feature scaling
+scaler = StandardScaler()
+X_scaled = scaler.fit_transform(X)
+
+"""Train the model :"""
+
+# Base model
+base_model = LogisticRegression()
+
+# Self-Training Wrapper
+self_training_model = SelfTrainingClassifier(base_model, criterion='k_best', k_best=3, max_iter=10)
+
+# Train on partially labeled data
+self_training_model.fit(X_scaled, y_semi)
+
+print("✅ Self-training complete!")
+
+"""Prediction :"""
+
+# Predict on all samples
+predicted = self_training_model.predict(X_scaled)
+
+# Add predictions to DataFrame
+df['predicted_attendance'] = predicted
+
+# Save model and predictions
+joblib.dump(self_training_model, "model.pkl")
+joblib.dump(scaler, "scaler.pkl")
+df.to_csv("predictions.csv", index=False)
+
+print("✅ Model and predictions saved as model.pkl and predictions.csv")
+
+"""Test the model:"""
+
+# Load saved model and scaler
+model = joblib.load("model.pkl")
+scaler = joblib.load("scaler.pkl")
+
+# 👇 Define your test input
+# Format: [interest (0-1), proximity (0-1), past_attendance (integer), free_time (0-1)]
+test_input = np.array([[0.0, 0.0, 0, 0.0]])
+  # You can change these values
+
+# Scale input the same way training data was scaled
+test_scaled = scaler.transform(test_input)
+
+# Make prediction
+prediction = model.predict(test_scaled)[0]
+
+# Show result
+if prediction == 1:
+    print("✅ The person is likely to ATTEND the event.")
+else:
+    print("❌ The person is NOT likely to attend the event.")
+
+"""Accuracy, Precision, Recall , F1(for classification):"""
+
+from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score
+
+# Filter only rows with actual labels (i.e., labeled data)
+labeled_mask = y_semi != -1
+X_labeled = X_scaled[labeled_mask]
+y_true = y_semi[labeled_mask]
+y_pred = self_training_model.predict(X_labeled)
+
+# Classification Metrics
+acc = accuracy_score(y_true, y_pred)
+prec = precision_score(y_true, y_pred)
+rec = recall_score(y_true, y_pred)
+f1 = f1_score(y_true, y_pred)
+
+# Display results
+print(f"📊 Model Evaluation on Labeled Data:")
+print(f"✅ Accuracy:  {acc:.4f}")
+print(f"✅ Precision: {prec:.4f}")
+print(f"✅ Recall:    {rec:.4f}")
+print(f"✅ F1 Score:  {f1:.4f}")
+
+import gradio as gr
+import joblib
+import numpy as np
+
+# Load model and scaler
+model = joblib.load("model.pkl")
+scaler = joblib.load("scaler.pkl")
+
+# Prediction function
+def predict_attendance(interest_pct, proximity_pct, past_attendance, free_time_pct):
+    # Convert % (0 to 100) to scale 1 to 2, then normalize to 0–1
+    interest = (interest_pct / 100)
+    proximity = (proximity_pct / 100)
+    free_time = (free_time_pct / 100)
+
+    # Prepare and scale input
+    input_data = np.array([[interest, proximity, past_attendance, free_time]])
+    input_scaled = scaler.transform(input_data)
+
+    # Predict
+    prediction = model.predict(input_scaled)[0]
+    return "✅ Will Attend" if prediction == 1 else "❌ Will Not Attend"
+
+# Gradio UI
+iface = gr.Interface(
+    fn=predict_attendance,
+    inputs=[
+        gr.Slider(0, 100, step=10, label="Interest in Topic (%)"),
+        gr.Slider(0, 100, step=10, label="Proximity to Event (%)"),
+        gr.Slider(0, 10, step=1, label="Similar Events Attended"),
+        gr.Slider(0, 100, step=10, label="Free Time Availability (%)"),
+    ],
+    outputs="text",
+    title="🎯 Event Attendance Predictor",
+    description="Enter your info to find out if you're likely to attend this event. Sliders use percent to indicate strength or availability."
+)
+
+iface.launch()

model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a49cf98b439fa45ec267bcfd68cb5f5c3d387de5f3bcfd0358b0866b4d06a7d
+size 2064

scaler.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6fabef7b51f48aa5fd1a94f31e52e8cd8d68e7f34dc9c3902ec23014a094bfe
+size 711