Upload folder using huggingface_hub

README.md

@@ -1,3 +1,14 @@
----
-license: apache-2.0
----
+# Mental Health Prediction Model
+## Overview
+This project predicts mental health status (Low, Moderate, High) using a Flask app and an XGBoost model.
+## Files
+- app.py: Flask app for predictions
+- index.html: Web interface
+- mental_health_model_final.pkl: Trained XGBoost model
+- scaler_final.pkl: Scaler for preprocessing
+- le_gender_final.pkl: Label encoder for Gender
+- le_target_final.pkl: Label encoder for target
+## How to Use
+1. Install dependencies: `pip install -r requirements.txt`
+2. Run the app: `python app.py`
+3. Access at `http://localhost:5000`

app.py

@@ -0,0 +1,92 @@
+from flask import Flask, request, jsonify, send_from_directory
+import pandas as pd
+import numpy as np
+import joblib
+import os
+
+app = Flask(__name__)
+
+# Load the model and preprocessing objects
+model = joblib.load('mental_health_model_final.pkl')
+scaler = joblib.load('scaler_final.pkl')
+le_gender = joblib.load('le_gender_final.pkl')
+le_target = joblib.load('le_target_final.pkl')
+
+# Print feature importance
+columns = ['Sentiment_Score', 'HRV', 'Sleep_Hours', 'Activity', 'Age', 'Gender', 'Work_Study_Hours']
+print("Feature Importance:")
+for feature, importance in zip(columns, model.feature_importances_):
+    print(f"{feature}: {importance:.4f}")
+
+# Serve the HTML file
+@app.route('/')
+def serve_html():
+    return send_from_directory('.', 'index.html')
+
+@app.route('/predict', methods=['POST'])
+def predict():
+    try:
+        # Get the input data from the request
+        data = request.get_json()
+        
+        # Convert to DataFrame
+        columns = ['Sentiment_Score', 'HRV', 'Sleep_Hours', 'Activity', 'Age', 'Gender', 'Work_Study_Hours']
+        new_data = pd.DataFrame([data], columns=columns)
+        
+        # Validate input
+        if new_data['Sleep_Hours'].iloc[0] < 0:
+            return jsonify({'error': 'Sleep_Hours cannot be negative'}), 400
+        if new_data['Work_Study_Hours'].iloc[0] < 0:
+            return jsonify({'error': 'Work_Study_Hours cannot be negative'}), 400
+        if new_data['Gender'].iloc[0] not in ['Male', 'Female']:
+            return jsonify({'error': 'Gender must be Male or Female'}), 400
+        
+        # Preprocess the data
+        new_data['Gender'] = le_gender.transform(new_data['Gender'])
+        new_data_scaled = scaler.transform(new_data)
+        
+        # Predict
+        prediction = model.predict(new_data_scaled)[0]
+        probs = model.predict_proba(new_data_scaled)[0]
+        health_status = {0: "Low", 1: "Moderate", 2: "High"}
+        result = health_status[prediction]
+        
+        # Include probabilities in the response
+        prob_dict = {health_status[i]: float(prob) for i, prob in enumerate(probs)}
+        
+        # Add a disclaimer for borderline predictions
+        max_prob = probs.max()
+        disclaimer = ""
+        if max_prob < 0.7:
+            disclaimer = "This prediction is uncertain (confidence below 70%). Please consult a professional for an accurate assessment."
+        
+        # Add a simple chatbot-like message
+        if result == "Low":
+            chatbot_message = "It looks like you might be experiencing low mental health. Consider reaching out to a friend or professional for support."
+        elif result == "Moderate":
+            chatbot_message = "Your mental health seems moderate. Keep up with self-care practices, and consider talking to someone if you feel overwhelmed."
+        else:  # High
+            chatbot_message = "Great news! Your mental health appears to be high. Keep maintaining your healthy habits!"
+
+        return jsonify({
+            'prediction': result,
+            'probabilities': prob_dict,
+            'disclaimer': disclaimer,
+            'model_accuracy': 'This model has a cross-validation accuracy of 76.8%.',
+            'chatbot_message': chatbot_message
+        })
+    except Exception as e:
+        return jsonify({'error': str(e)}), 400
+
+@app.route('/feedback', methods=['POST'])
+def feedback():
+    try:
+        feedback_data = request.get_json()
+        with open('feedback.log', 'a') as f:
+            f.write(f"Prediction: {feedback_data['prediction']}, Accurate: {feedback_data['accurate']}\n")
+        return jsonify({'message': 'Feedback received'})
+    except Exception as e:
+        return jsonify({'error': str(e)}), 400
+
+if __name__ == '__main__':
+    app.run(debug=True, host='0.0.0.0', port=5000)

index.html

@@ -0,0 +1,224 @@
+<!DOCTYPE html>
+<html>
+<head>
+    <title>Mental Health Prediction</title>
+    <style>
+        body {
+            font-family: Arial, sans-serif;
+            margin: 0;
+            padding: 0;
+            display: flex;
+            justify-content: center;
+            align-items: center;
+            height: 100vh;
+            background: linear-gradient(to right, #36d1dc, #5b86e5);
+            animation: fadeIn 1s ease-in-out;
+        }
+        @keyframes fadeIn {
+            from { opacity: 0; }
+            to { opacity: 1; }
+        }
+        .container {
+            background: white;
+            padding: 20px;
+            border-radius: 10px;
+            box-shadow: 0 10px 20px rgba(0, 0, 0, 0.2);
+            text-align: center;
+            width: 400px;
+            animation: slideUp 0.8s ease-in-out;
+        }
+        @keyframes slideUp {
+            from { transform: translateY(30px); opacity: 0; }
+            to { transform: translateY(0); opacity: 1; }
+        }
+        h2 {
+            color: #333;
+        }
+        label {
+            display: block;
+            text-align: left;
+            margin: 5px 0 2px;
+            color: #555;
+        }
+        input, select, button {
+            width: 100%;
+            padding: 10px;
+            margin: 5px 0 10px;
+            border: 1px solid #ccc;
+            border-radius: 5px;
+            transition: 0.3s;
+            box-sizing: border-box;
+        }
+        input:focus, select:focus {
+            border-color: #5b86e5;
+            box-shadow: 0 0 8px rgba(91, 134, 229, 0.6);
+        }
+        button {
+            background: #007bff;
+            color: white;
+            cursor: pointer;
+            border: none;
+            font-weight: bold;
+            transition: 0.3s;
+        }
+        button:hover {
+            background: #0056b3;
+            transform: scale(1.05);
+        }
+        #result, #probabilities, #disclaimer, #model-accuracy {
+            margin-top: 10px;
+            color: #333;
+        }
+        #result {
+            font-weight: bold;
+            color: #007bff;
+        }
+        #probabilities table {
+            width: 100%;
+            margin-top: 10px;
+            border-collapse: collapse;
+        }
+        #probabilities th, #probabilities td {
+            padding: 5px;
+            border: 1px solid #ccc;
+            text-align: center;
+        }
+        #probabilities th {
+            background: #f0f0f0;
+        }
+        #disclaimer {
+            font-style: italic;
+            color: #e74c3c;
+        }
+        #model-accuracy {
+            font-size: 0.9em;
+            color: #555;
+        }
+    </style>
+    <script>
+        function predictMentalHealth() {
+            // Collect form data
+            const form = document.getElementById('predictionForm');
+            const formData = new FormData(form);
+            
+            // Convert form data to JSON
+            const data = {
+                Sentiment_Score: parseFloat(formData.get('Sentiment_Score')),
+                HRV: parseFloat(formData.get('HRV')),
+                Sleep_Hours: parseFloat(formData.get('Sleep_Hours')),
+                Activity: parseInt(formData.get('Activity')),
+                Age: parseInt(formData.get('Age')),
+                Gender: formData.get('Gender'),
+                Work_Study_Hours: parseFloat(formData.get('Work_Study_Hours'))
+            };
+            
+            // Validate input
+            if (isNaN(data.Sentiment_Score) || data.Sentiment_Score < 0 || data.Sentiment_Score > 1) {
+                document.getElementById('result').innerText = 'Error: Sentiment Score must be between 0 and 1';
+                return;
+            }
+            if (isNaN(data.HRV) || data.HRV < 0) {
+                document.getElementById('result').innerText = 'Error: HRV cannot be negative';
+                return;
+            }
+            if (isNaN(data.Sleep_Hours) || data.Sleep_Hours < 0) {
+                document.getElementById('result').innerText = 'Error: Sleep Hours cannot be negative';
+                return;
+            }
+            if (isNaN(data.Activity) || data.Activity < 0) {
+                document.getElementById('result').innerText = 'Error: Activity cannot be negative';
+                return;
+            }
+            if (isNaN(data.Age) || data.Age < 0) {
+                document.getElementById('result').innerText = 'Error: Age cannot be negative';
+                return;
+            }
+            if (isNaN(data.Work_Study_Hours) || data.Work_Study_Hours < 0) {
+                document.getElementById('result').innerText = 'Error: Work/Study Hours cannot be negative';
+                return;
+            }
+            
+            // Send request to the Flask API
+            fetch('/predict', {
+                method: 'POST',
+                headers: {
+                    'Content-Type': 'application/json'
+                },
+                body: JSON.stringify(data)
+            })
+            .then(response => response.json())
+            .then(data => {
+                if (data.prediction) {
+                    // Display prediction
+                    document.getElementById('result').innerText = 'Prediction: ' + data.prediction;
+                    
+                    // Display probabilities in a table
+                    const probsTable = document.getElementById('probabilities-table');
+                    probsTable.innerHTML = '';
+                    for (const [status, prob] of Object.entries(data.probabilities)) {
+                        const row = probsTable.insertRow();
+                        row.insertCell(0).innerText = status;
+                        row.insertCell(1).innerText = (prob * 100).toFixed(2) + '%';
+                    }
+                    
+                    // Display disclaimer and model accuracy
+                    document.getElementById('disclaimer').innerText = data.disclaimer || '';
+                    document.getElementById('model-accuracy').innerText = data.model_accuracy;
+                } else {
+                    document.getElementById('result').innerText = 'Error: ' + data.error;
+                    document.getElementById('probabilities-table').innerHTML = '';
+                    document.getElementById('disclaimer').innerText = '';
+                    document.getElementById('model-accuracy').innerText = '';
+                }
+            })
+            .catch(error => {
+                document.getElementById('result').innerText = 'Request failed';
+                document.getElementById('probabilities-table').innerHTML = '';
+                document.getElementById('disclaimer').innerText = '';
+                document.getElementById('model-accuracy').innerText = '';
+                console.error('Error:', error);
+            });
+        }
+    </script>
+</head>
+<body>
+    <div class="container">
+        <h2>Mental Health Prediction</h2>
+        <form id="predictionForm" onsubmit="event.preventDefault(); predictMentalHealth();">
+            <label>Sentiment Score (0 to 1):</label>
+            <input type="number" step="0.01" name="Sentiment_Score" required>
+            
+            <label>HRV (Heart Rate Variability):</label>
+            <input type="number" step="0.01" name="HRV" required>
+            
+            <label>Sleep Hours:</label>
+            <input type="number" step="0.1" name="Sleep_Hours" required>
+            
+            <label>Activity (Steps/Day):</label>
+            <input type="number" name="Activity" required>
+            
+            <label>Age:</label>
+            <input type="number" name="Age" required>
+            
+            <label>Gender:</label>
+            <select name="Gender" required>
+                <option value="Male">Male</option>
+                <option value="Female">Female</option>
+            </select>
+            
+            <label>Work/Study Hours:</label>
+            <input type="number" step="0.1" name="Work_Study_Hours" required>
+            
+            <button type="submit">Predict</button>
+        </form>
+        <h3 id="result"></h3>
+        <div id="probabilities">
+            <table id="probabilities-table">
+                <tr><th>Status</th><th>Probability</th></tr>
+            </table>
+        </div>
+        <div id="disclaimer"></div>
+        <div id="model-accuracy"></div>
+    </div>
+</body>
+</html>

le_gender_final.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae0c50f426a66d9e70886753e66e2493f723d9ca80c53709ea7796ce47eac18a
+size 543

le_target_final.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:356dcc80ba23896d339529d04c0b7cda36bedd28f3ab9d56661f5c6433135eb1
+size 555

mental_health_model_final.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9003f5d318ac9071a886e2af9384936b511bf4417a8969e595b67c05fa76db05
+size 1261470

scaler_final.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9dbdce9b70795ca2c1d9396f541c96e5084971eb9d9d5138e3898b97112587e
+size 1167

upload_to_hf.py

@@ -0,0 +1,16 @@
+from huggingface_hub import HfApi
+
+# Initialize the API (will use the token from CLI authentication)
+api = HfApi()
+
+# Define your repository ID
+repo_id = "AshutoshAI/mental-health-predictor"
+
+# Upload the entire folder, excluding unnecessary files
+api.upload_folder(
+    folder_path="D:/DSMINIPRO",
+    repo_id=repo_id,
+    repo_type="model",
+    ignore_patterns=["*.csv", "*le_gender.pkl", "*le_target.pkl", "*mental_health_model.pkl", "*scaler.pkl", "*scaler_updated.pkl", "*.log"],
+)
+print(f"Successfully uploaded to {repo_id}")