Add exoplanet classification Gradio app with LFS

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.pem filter=lfs diff=lfs merge=lfs -text

.gradio/certificate.pem

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c99356c265ee06c0ae0502e74d38231263513726d001cfe28ea25e70af2cc7f
+size 1970

app.py

@@ -0,0 +1,818 @@
+"""
+Exoplanet Classification Web Application
+==========================================
+Modern Gradio interface for TESS exoplanet classification
+Features individual parameter inputs and batch processing
+
+Author: ML Pipeline
+Date: October 4, 2025
+"""
+
+import warnings
+warnings.filterwarnings('ignore')
+
+import gradio as gr
+import pandas as pd
+import numpy as np
+import joblib
+import json
+from pathlib import Path
+from datetime import datetime
+import traceback
+
+# ====================================================================
+# CONFIGURATION
+# ====================================================================
+
+MODEL_DIR = Path('mutiresult')
+
+BASE_FEATURES = [
+    "ra", "dec", "st_pmra", "st_pmraerr1", "st_pmraerr2", "st_pmralim",
+    "st_pmdec", "st_pmdecerr1", "st_pmdecerr2", "st_pmdeclim",
+    "pl_tranmid", "pl_tranmiderr1", "pl_tranmiderr2", "pl_tranmidlim",
+    "pl_orbper", "pl_orbpererr1", "pl_orbpererr2", "pl_orbperlim",
+    "pl_trandurh", "pl_trandurherr1", "pl_trandurherr2", "pl_trandurhlim",
+    "pl_trandep", "pl_trandeperr1", "pl_trandeperr2", "pl_trandeplim",
+    "pl_rade", "pl_radeerr1", "pl_radeerr2", "pl_radelim",
+    "pl_insol", "pl_eqt",
+    "st_tmag", "st_tmagerr1", "st_tmagerr2", "st_tmaglim",
+    "st_dist", "st_disterr1", "st_disterr2", "st_distlim",
+    "st_teff", "st_tefferr1", "st_tefferr2", "st_tefflim",
+    "st_logg", "st_loggerr1", "st_loggerr2", "st_logglim",
+    "st_rad", "st_raderr1", "st_raderr2", "st_radlim"
+]
+
+CLASS_DESCRIPTIONS = {
+    'APC': 'Ambiguous Planet Candidate',
+    'CP': 'Confirmed Planet',
+    'FA': 'False Alarm',
+    'FP': 'False Positive',
+    'KP': 'Known Planet',
+    'PC': 'Planet Candidate'
+}
+
+# Comprehensive parameter descriptions
+PARAM_INFO = {
+    'ra': ('Right Ascension (RA)', '0-360°. Celestial longitude coordinate of the star'),
+    'dec': ('Declination (Dec)', '-90 to +90°. Celestial latitude coordinate of the star'),
+    'st_pmra': ('Proper Motion in RA', 'mas/yr. Star movement in RA direction'),
+    'st_pmraerr1': ('PM RA Upper Error', 'mas/yr. Positive uncertainty'),
+    'st_pmraerr2': ('PM RA Lower Error', 'mas/yr. Negative uncertainty'),
+    'st_pmralim': ('PM RA Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'st_pmdec': ('Proper Motion in Dec', 'mas/yr. Star movement in Dec direction'),
+    'st_pmdecerr1': ('PM Dec Upper Error', 'mas/yr. Positive uncertainty'),
+    'st_pmdecerr2': ('PM Dec Lower Error', 'mas/yr. Negative uncertainty'),
+    'st_pmdeclim': ('PM Dec Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'pl_tranmid': ('Transit Midpoint Time', 'BJD. Time when planet crosses star center'),
+    'pl_tranmiderr1': ('Transit Mid Upper Error', 'days. Positive uncertainty'),
+    'pl_tranmiderr2': ('Transit Mid Lower Error', 'days. Negative uncertainty'),
+    'pl_tranmidlim': ('Transit Mid Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'pl_orbper': ('Orbital Period', 'days. Time for one complete orbit'),
+    'pl_orbpererr1': ('Orbital Period Upper Error', 'days. Positive uncertainty'),
+    'pl_orbpererr2': ('Orbital Period Lower Error', 'days. Negative uncertainty'),
+    'pl_orbperlim': ('Orbital Period Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'pl_trandurh': ('Transit Duration', 'hours. How long planet blocks star'),
+    'pl_trandurherr1': ('Transit Duration Upper Error', 'hours. Positive uncertainty'),
+    'pl_trandurherr2': ('Transit Duration Lower Error', 'hours. Negative uncertainty'),
+    'pl_trandurhlim': ('Transit Duration Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'pl_trandep': ('Transit Depth', '%. Fraction of starlight blocked'),
+    'pl_trandeperr1': ('Transit Depth Upper Error', '%. Positive uncertainty'),
+    'pl_trandeperr2': ('Transit Depth Lower Error', '%. Negative uncertainty'),
+    'pl_trandeplim': ('Transit Depth Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'pl_rade': ('Planet Radius', 'Earth radii (R⊕). Planet size vs Earth'),
+    'pl_radeerr1': ('Planet Radius Upper Error', 'R⊕. Positive uncertainty'),
+    'pl_radeerr2': ('Planet Radius Lower Error', 'R⊕. Negative uncertainty'),
+    'pl_radelim': ('Planet Radius Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'pl_insol': ('Insolation Flux', 'Earth flux. Stellar energy hitting planet'),
+    'pl_eqt': ('Equilibrium Temperature', 'Kelvin. Expected planet surface temperature'),
+    'st_tmag': ('TESS Magnitude', 'mag. Star brightness (lower=brighter, typical 6-16)'),
+    'st_tmagerr1': ('TESS Mag Upper Error', 'mag. Positive uncertainty'),
+    'st_tmagerr2': ('TESS Mag Lower Error', 'mag. Negative uncertainty'),
+    'st_tmaglim': ('TESS Mag Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'st_dist': ('Distance', 'parsecs. Distance to star (1 pc ≈ 3.26 light-years)'),
+    'st_disterr1': ('Distance Upper Error', 'parsecs. Positive uncertainty'),
+    'st_disterr2': ('Distance Lower Error', 'parsecs. Negative uncertainty'),
+    'st_distlim': ('Distance Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'st_teff': ('Effective Temperature', 'Kelvin. Star surface temp (Sun ≈ 5778K)'),
+    'st_tefferr1': ('Effective Temp Upper Error', 'K. Positive uncertainty'),
+    'st_tefferr2': ('Effective Temp Lower Error', 'K. Negative uncertainty'),
+    'st_tefflim': ('Effective Temp Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'st_logg': ('Surface Gravity', 'log₁₀(cm/s²). Stellar gravity (Sun ≈ 4.44)'),
+    'st_loggerr1': ('Surface Gravity Upper Error', 'dex. Positive uncertainty'),
+    'st_loggerr2': ('Surface Gravity Lower Error', 'dex. Negative uncertainty'),
+    'st_logglim': ('Surface Gravity Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+    'st_rad': ('Stellar Radius', 'Solar radii (R☉). Star size vs Sun'),
+    'st_raderr1': ('Stellar Radius Upper Error', 'R☉. Positive uncertainty'),
+    'st_raderr2': ('Stellar Radius Lower Error', 'R☉. Negative uncertainty'),
+    'st_radlim': ('Stellar Radius Limit Flag', '0=measured, 1=upper limit, -1=lower limit'),
+}
+
+# ====================================================================
+# MODEL LOADING
+# ====================================================================
+
+class ModelPipeline:
+    def __init__(self, model_dir):
+        self.model_dir = Path(model_dir)
+        self.models = {}
+        self.scaler = None
+        self.imputer = None
+        self.label_encoder = None
+        self.feature_info = None
+        self.metrics = None
+        self._load_artifacts()
+    
+    def _load_artifacts(self):
+        try:
+            print("Loading preprocessing artifacts...")
+            self.scaler = joblib.load(self.model_dir / 'scaler.pkl')
+            self.imputer = joblib.load(self.model_dir / 'imputer.pkl')
+            self.label_encoder = joblib.load(self.model_dir / 'label_encoder.pkl')
+            
+            with open(self.model_dir / 'feature_info.json', 'r') as f:
+                self.feature_info = json.load(f)
+            
+            with open(self.model_dir / 'metrics_summary.json', 'r') as f:
+                self.metrics = json.load(f)
+            
+            print("Loading trained models...")
+            model_files = {
+                'LightGBM': 'model_lightgbm.pkl',
+                'XGBoost': 'model_xgboost.pkl',
+                'RandomForest': 'model_randomforest.pkl',
+                'ExtraTrees': 'model_extratrees.pkl',
+                'LogisticRegression': 'model_logisticregression.pkl'
+            }
+            
+            for name, filename in model_files.items():
+                model_path = self.model_dir / filename
+                if model_path.exists():
+                    self.models[name] = joblib.load(model_path)
+                    print(f"  ✓ Loaded {name}")
+            
+            print(f"\n✓ Successfully loaded {len(self.models)} models")
+            print(f"✓ Target classes: {list(self.label_encoder.classes_)}")
+            
+        except Exception as e:
+            print(f"Error loading artifacts: {str(e)}")
+            raise
+    
+    def create_features(self, df):
+        X = df.copy()
+        
+        if 'st_rad' in X.columns and 'st_teff' in X.columns:
+            X['luminosity_proxy'] = X['st_rad']**2 * X['st_teff']**4
+        
+        if 'pl_rade' in X.columns and 'st_rad' in X.columns:
+            X['planet_star_radius_ratio'] = X['pl_rade'] / (X['st_rad'] + 1e-10)
+            X['transit_depth_proxy'] = (X['pl_rade'] / (X['st_rad'] + 1e-10))**2
+        
+        if 'st_dist' in X.columns:
+            X['log_distance'] = np.log1p(X['st_dist'])
+            X['inv_distance'] = 1 / (X['st_dist'] + 1e-10)
+        
+        if 'pl_eqt' in X.columns and 'st_teff' in X.columns:
+            X['temp_ratio'] = X['pl_eqt'] / (X['st_teff'] + 1e-10)
+        
+        if 'pl_insol' in X.columns:
+            X['log_insol'] = np.log1p(X['pl_insol'])
+            X['sqrt_insol'] = np.sqrt(X['pl_insol'] + 1e-10)
+        
+        if 'pl_orbper' in X.columns:
+            X['log_orbper'] = np.log1p(X['pl_orbper'])
+        
+        error_pairs = [
+            ('st_pmra', 'st_pmraerr1'),
+            ('st_pmdec', 'st_pmdecerr1'),
+            ('pl_rade', 'pl_radeerr1'),
+        ]
+        
+        for base, error in error_pairs:
+            if base in X.columns and error in X.columns:
+                X[f'{base}_error_ratio'] = np.abs(X[error]) / (np.abs(X[base]) + 1e-10)
+        
+        numeric_cols = [col for col in BASE_FEATURES[:10] if col in X.columns]
+        if len(numeric_cols) > 3:
+            X['feature_mean'] = X[numeric_cols].mean(axis=1)
+            X['feature_std'] = X[numeric_cols].std(axis=1)
+            X['feature_max'] = X[numeric_cols].max(axis=1)
+            X['feature_min'] = X[numeric_cols].min(axis=1)
+        
+        X = X.replace([np.inf, -np.inf], np.nan)
+        for col in X.columns:
+            if X[col].isnull().any():
+                X[col].fillna(X[col].median(), inplace=True)
+        
+        return X
+    
+    def preprocess(self, df):
+        """Preprocess with correct order: impute → engineer → scale → select"""
+        # Step 1: Apply imputer to base 52 features
+        imputer_features = list(self.imputer.feature_names_in_)
+        X_base = df[imputer_features].copy()
+        X_imputed = self.imputer.transform(X_base)
+        X_imputed_df = pd.DataFrame(X_imputed, columns=imputer_features, index=df.index)
+        
+        # Step 2: Create engineered features (from 52 to ~68 features)
+        X_engineered = self.create_features(X_imputed_df)
+        
+        # Step 3: Apply scaler to ALL engineered features (68 features)
+        # IMPORTANT: Scaler was fitted on all engineered features, not selected ones
+        X_scaled = self.scaler.transform(X_engineered)
+        X_scaled_df = pd.DataFrame(X_scaled, columns=X_engineered.columns, index=df.index)
+        
+        # Step 4: Select final 60 features AFTER scaling
+        X_selected = X_scaled_df[self.feature_info['selected_features']]
+        
+        return X_selected
+    
+    def predict(self, df, model_name='Ensemble'):
+        X = self.preprocess(df)
+        
+        if model_name == 'Ensemble':
+            probabilities = []
+            weights = [1.5, 1.5, 1.0, 1.0]
+            models_to_use = ['LightGBM', 'XGBoost', 'RandomForest', 'ExtraTrees']
+            
+            for model_name_single in models_to_use:
+                if model_name_single in self.models:
+                    proba = self.models[model_name_single].predict_proba(X)
+                    probabilities.append(proba)
+            
+            avg_proba = np.average(probabilities, axis=0, weights=weights[:len(probabilities)])
+            y_pred = np.argmax(avg_proba, axis=1)
+            predictions = self.label_encoder.inverse_transform(y_pred)
+            
+            return predictions, avg_proba
+        else:
+            model = self.models[model_name]
+            y_pred = model.predict(X)
+            y_proba = model.predict_proba(X)
+            predictions = self.label_encoder.inverse_transform(y_pred)
+            
+            return predictions, y_proba
+
+
+print("Initializing Model Pipeline...")
+pipeline = ModelPipeline(MODEL_DIR)
+
+# ====================================================================
+# PREDICTION FUNCTIONS
+# ====================================================================
+
+def predict_single(file, model_choice):
+    """Predict from single sample CSV file (must have exactly 2 rows: header + 1 data row)"""
+    try:
+        if file is None:
+            return "⚠️ Please upload a CSV file with 1 sample (2 rows: header + data)", "", ""
+        
+        df = pd.read_csv(file.name)
+        
+        # Validate: must have exactly 1 data row
+        if len(df) == 0:
+            error_html = """
+            <div style="background: #ffebee; border-left: 5px solid #f44336; padding: 20px; border-radius: 10px;">
+                <h3 style="color: #c62828; margin-top: 0;">❌ Empty File</h3>
+                <p style="color: #c62828;">CSV file has no data rows. Please upload a file with exactly 1 data row (2 total rows including header).</p>
+            </div>
+            """
+            return error_html, "", ""
+        
+        if len(df) > 1:
+            error_html = f"""
+            <div style="background: #fff3e0; border-left: 5px solid #ff9800; padding: 20px; border-radius: 10px;">
+                <h3 style="color: #e65100; margin-top: 0;">⚠️ Too Many Rows</h3>
+                <p style="color: #e65100;">
+                    Your CSV file has <strong>{len(df)} data rows</strong>. For single sample prediction, the file must have exactly <strong>1 data row</strong> (2 total rows: header + data).
+                </p>
+                <p style="color: #e65100;">
+                    If you want to process multiple samples, please use the <strong>"📦 Batch Processing"</strong> tab instead.
+                </p>
+            </div>
+            """
+            return error_html, "", ""
+        
+        # Validate columns
+        missing_cols = [col for col in BASE_FEATURES if col not in df.columns]
+        if len(missing_cols) > 0:
+            error_html = f"""
+            <div style="background: #ffebee; border-left: 5px solid #f44336; padding: 20px; border-radius: 10px;">
+                <h3 style="color: #c62828; margin-top: 0;">❌ Missing Required Columns</h3>
+                <p style="color: #c62828;">Your CSV is missing <strong>{len(missing_cols)}</strong> required columns:</p>
+                <pre style="background: #fff; padding: 15px; border-radius: 5px; overflow-x: auto; color: #c62828;">
+{', '.join(missing_cols[:30])}
+                </pre>
+                <p style="color: #c62828;">Please ensure all 52 base features are present. Download the template for reference.</p>
+            </div>
+            """
+            return error_html, "", ""
+        
+        # Get ID if exists
+        id_value = "N/A"
+        for possible_id in ['id', 'ID', 'toi', 'TOI', 'tid', 'TID']:
+            if possible_id in df.columns:
+                id_value = str(df[possible_id].iloc[0])
+                break
+        
+        # Make prediction
+        predictions, probabilities = pipeline.predict(df, model_name=model_choice)
+        
+        pred_label = predictions[0]
+        pred_confidence = probabilities[0].max()
+        
+        # Beautiful result card
+        result_html = f"""
+        <div style="background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); 
+                    padding: 40px; border-radius: 20px; box-shadow: 0 20px 60px rgba(0,0,0,0.3);">
+            <div style="text-align: center; color: white;">
+                <h1 style="margin: 0 0 30px 0; font-size: 42px; text-shadow: 2px 2px 4px rgba(0,0,0,0.3);">
+                    🎯 Classification Result
+                </h1>
+                <div style="background: rgba(255,255,255,0.15); padding: 40px; border-radius: 15px; 
+                            backdrop-filter: blur(10px); border: 2px solid rgba(255,255,255,0.2);">
+                    <div style="margin-bottom: 30px;">
+                        <div style="font-size: 18px; opacity: 0.9; margin-bottom: 10px;">Sample ID: {id_value}</div>
+                        <div style="font-size: 22px; opacity: 0.95; margin-bottom: 15px; letter-spacing: 2px;">
+                            PREDICTED CLASS
+                        </div>
+                        <div style="font-size: 72px; font-weight: bold; margin: 20px 0; 
+                                    text-shadow: 3px 3px 6px rgba(0,0,0,0.4); letter-spacing: 4px;">
+                            {pred_label}
+                        </div>
+                        <div style="font-size: 20px; opacity: 0.9; font-style: italic;">
+                            {CLASS_DESCRIPTIONS.get(pred_label, '')}
+                        </div>
+                    </div>
+                    <div style="padding-top: 30px; border-top: 2px solid rgba(255,255,255,0.3);">
+                        <div style="font-size: 18px; opacity: 0.95; margin-bottom: 15px; letter-spacing: 1px;">
+                            CONFIDENCE SCORE
+                        </div>
+                        <div style="font-size: 54px; font-weight: bold; text-shadow: 2px 2px 4px rgba(0,0,0,0.3);">
+                            {pred_confidence:.1%}
+                        </div>
+                    </div>
+                </div>
+                <div style="margin-top: 25px; font-size: 16px; opacity: 0.85;">
+                    Model: {model_choice} | Accuracy: {pipeline.metrics[model_choice]['Accuracy']:.1%}
+                </div>
+            </div>
+        </div>
+        """
+        
+        # Probability bars
+        prob_html = "<div style='padding: 20px;'>"
+        prob_list = [(pipeline.label_encoder.classes_[i], probabilities[0, i]) 
+                     for i in range(len(pipeline.label_encoder.classes_))]
+        prob_list.sort(key=lambda x: x[1], reverse=True)
+        
+        for class_name, prob in prob_list:
+            color = "#4caf50" if class_name == pred_label else "#2196F3"
+            prob_html += f"""
+            <div style="margin-bottom: 20px;">
+                <div style="display: flex; justify-content: space-between; margin-bottom: 8px;">
+                    <span style="font-weight: bold; font-size: 18px;">{class_name}</span>
+                    <span style="font-weight: bold; font-size: 18px; color: {color};">{prob:.1%}</span>
+                </div>
+                <div style="font-size: 14px; color: #666; margin-bottom: 8px;">
+                    {CLASS_DESCRIPTIONS.get(class_name, '')}
+                </div>
+                <div style="background: #e0e0e0; border-radius: 10px; height: 30px; overflow: hidden;">
+                    <div style="background: linear-gradient(90deg, {color}, {color}aa); 
+                                width: {prob*100}%; height: 100%; border-radius: 10px;
+                                transition: width 0.3s ease; display: flex; align-items: center; 
+                                justify-content: flex-end; padding-right: 10px; color: white; 
+                                font-weight: bold; font-size: 14px;">
+                    </div>
+                </div>
+            </div>
+            """
+        prob_html += "</div>"
+        
+        # Input data preview
+        preview_html = f"""
+        <div style="background: #f5f5f5; padding: 20px; border-radius: 10px; margin-top: 20px;">
+            <h3 style="color: #333; margin-top: 0;">📋 Input Data Summary</h3>
+            <p style="color: #666;"><strong>Sample ID:</strong> {id_value}</p>
+            <p style="color: #666;"><strong>Features:</strong> {len(df.columns)} columns loaded</p>
+            <p style="color: #666;"><strong>Model:</strong> {model_choice}</p>
+        </div>
+        """
+        
+        return result_html, prob_html, preview_html
+        
+    except Exception as e:
+        error_html = f"""
+        <div style="background: #ffebee; border-left: 5px solid #f44336; padding: 20px; border-radius: 10px;">
+            <h3 style="color: #c62828; margin-top: 0;">❌ Prediction Error</h3>
+            <p style="color: #c62828;">{str(e)}</p>
+            <pre style="background: #fff; padding: 10px; border-radius: 5px; overflow-x: auto;">
+{traceback.format_exc()}
+            </pre>
+        </div>
+        """
+        return error_html, "", ""
+
+
+def predict_batch(file, model_choice):
+    """Batch prediction from CSV file"""
+    try:
+        if file is None:
+            return "⚠️ Please upload a CSV file", None, None
+        
+        df = pd.read_csv(file.name)
+        
+        if len(df) == 0:
+            return "⚠️ CSV file is empty", None, None
+        
+        # Validate columns
+        missing_cols = [col for col in BASE_FEATURES if col not in df.columns]
+        if len(missing_cols) > 0:
+            error_msg = f"""## ❌ Missing Required Columns
+
+Your CSV is missing {len(missing_cols)} required columns:
+
+```
+{', '.join(missing_cols[:20])}
+```
+
+Please ensure all 52 base features are present in your CSV file."""
+            return error_msg, None, None
+        
+        # Get ID column
+        id_col = None
+        for possible_id in ['id', 'ID', 'toi', 'TOI', 'tid', 'TID']:
+            if possible_id in df.columns:
+                id_col = possible_id
+                break
+        
+        # Predict
+        predictions, probabilities = pipeline.predict(df, model_name=model_choice)
+        
+        # Simple results
+        results = pd.DataFrame()
+        results['ID'] = df[id_col].values if id_col else range(1, len(df) + 1)
+        results['Label'] = predictions
+        
+        # Detailed results
+        detailed_results = results.copy()
+        detailed_results['Confidence'] = probabilities.max(axis=1)
+        for i, class_name in enumerate(pipeline.label_encoder.classes_):
+            detailed_results[f'Prob_{class_name}'] = probabilities[:, i]
+        
+        # Summary
+        summary = f"""## ✅ Batch Processing Complete
+
+**Model**: {model_choice}  
+**Samples Processed**: {len(df)}  
+**Time**: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}
+
+### Class Distribution:
+"""
+        for class_name in pipeline.label_encoder.classes_:
+            count = (predictions == class_name).sum()
+            pct = count / len(predictions) * 100
+            summary += f"\n- **{class_name}**: {count} ({pct:.1f}%)"
+        
+        summary += f"\n\n**Average Confidence**: {probabilities.max(axis=1).mean():.1%}"
+        
+        # Save file
+        timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
+        output_file = f'predictions_{timestamp}.csv'
+        results.to_csv(output_file, index=False)
+        
+        return summary, detailed_results, output_file
+        
+    except Exception as e:
+        return f"❌ Error: {str(e)}", None, None
+
+
+def get_template():
+    """Generate template CSV"""
+    template_df = pd.DataFrame(columns=['id'] + BASE_FEATURES)
+    template_df.loc[0] = [1] + [0.0] * len(BASE_FEATURES)
+    template_file = 'template_exoplanet.csv'
+    template_df.to_csv(template_file, index=False)
+    return template_file
+
+
+# ====================================================================
+# GRADIO UI
+# ====================================================================
+
+custom_css = """
+.gradio-container {
+    font-family: 'Segoe UI', Arial, sans-serif !important;
+    max-width: 1800px !important;
+}
+.gr-button-primary {
+    background: linear-gradient(90deg, #667eea 0%, #764ba2 100%) !important;
+    border: none !important;
+    font-weight: 600 !important;
+    font-size: 18px !important;
+}
+.gr-button-secondary {
+    background: linear-gradient(90deg, #4caf50 0%, #45a049 100%) !important;
+    border: none !important;
+    color: white !important;
+}
+"""
+
+with gr.Blocks(css=custom_css, title="🌍 Exoplanet Classifier", theme=gr.themes.Soft()) as app:
+    
+    # Header
+    gr.HTML("""
+    <div style="text-align: center; background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); 
+                padding: 50px; border-radius: 20px; margin-bottom: 30px; box-shadow: 0 10px 30px rgba(0,0,0,0.2);">
+        <h1 style="color: white; font-size: 56px; margin: 0; text-shadow: 2px 2px 4px rgba(0,0,0,0.3);">
+            🌍 TESS Exoplanet Classifier
+        </h1>
+        <p style="color: rgba(255,255,255,0.95); font-size: 24px; margin: 20px 0 0 0;">
+            AI-Powered Classification System for Transiting Exoplanet Survey Satellite Data
+        </p>
+        <div style="margin-top: 20px; display: flex; justify-content: center; gap: 30px; flex-wrap: wrap;">
+            <div style="background: rgba(255,255,255,0.2); padding: 15px 30px; border-radius: 10px;">
+                <div style="font-size: 32px; font-weight: bold;">72.2%</div>
+                <div style="font-size: 14px; opacity: 0.9;">Ensemble Accuracy</div>
+            </div>
+            <div style="background: rgba(255,255,255,0.2); padding: 15px 30px; border-radius: 10px;">
+                <div style="font-size: 32px; font-weight: bold;">6</div>
+                <div style="font-size: 14px; opacity: 0.9;">Target Classes</div>
+            </div>
+            <div style="background: rgba(255,255,255,0.2); padding: 15px 30px; border-radius: 10px;">
+                <div style="font-size: 32px; font-weight: bold;">5</div>
+                <div style="font-size: 14px; opacity: 0.9;">ML Models</div>
+            </div>
+        </div>
+    </div>
+    """)
+    
+    # Info
+    with gr.Accordion("📖 About & Instructions", open=False):
+        gr.Markdown("""
+        ## 🎯 Classification Classes
+        
+        | Code | Name | Description |
+        |------|------|-------------|
+        | **APC** | Ambiguous Planet Candidate | Requires further analysis |
+        | **CP** | Confirmed Planet | Validated exoplanet |
+        | **FA** | False Alarm | Instrumental/systematic effects |
+        | **FP** | False Positive | Astrophysical false positive |
+        | **KP** | Known Planet | Previously discovered |
+        | **PC** | Planet Candidate | High-confidence candidate |
+        
+        ## 📊 Models
+        
+        - **Ensemble**: Weighted combination (72.2% accuracy) - **RECOMMENDED**
+        - **XGBoost**: Gradient boosting (72.2%)
+        - **LightGBM**: Fast gradient boosting (71.6%)
+        - **Random Forest**: Tree ensemble (70.3%)
+        - **Extra Trees**: Randomized trees (70.8%)
+        
+        ## 💡 Usage
+        
+        **Single Sample**: Upload a CSV file with exactly 2 rows (header + 1 data row) containing all 52 required feature columns.
+        **Batch Processing**: Upload a CSV with all 52 required feature columns for bulk predictions of multiple samples.
+        """)
+    
+    # Tabs
+    with gr.Tabs():
+        
+        # ===== SINGLE SAMPLE TAB =====
+        with gr.Tab("🔬 Single Sample Prediction"):
+            gr.HTML("""
+            <div style="background: linear-gradient(to right, #e3f2fd, #f5f5f5); padding: 30px; 
+                        border-radius: 15px; border-left: 5px solid #2196F3; margin-bottom: 25px; 
+                        box-shadow: 0 4px 6px rgba(0,0,0,0.1);">
+                <h2 style="color: black; margin: 0 0 15px 0; font-size: 28px;">📋 Single Sample CSV Upload</h2>
+                <p style="color: black; margin: 0 0 10px 0; font-size: 17px; line-height: 1.7;">
+                    Upload a CSV file containing <strong style="color: black;">exactly 1 sample</strong> for classification.
+                </p>
+                <div style="background: rgba(33, 150, 243, 0.1); padding: 15px; border-radius: 8px; margin-top: 15px;">
+                    <p style="color: black; margin: 0; font-size: 15px;"><strong style="color: black;">📌 Important:</strong></p>
+                    <ul style="color: black; margin: 10px 0 0 0; padding-left: 20px;">
+                        <li style="color: black;">File must have <strong style="color: black;">exactly 2 rows</strong>: 1 header row + 1 data row</li>
+                        <li style="color: black;">Must contain all <strong style="color: black;">52 required feature columns</strong></li>
+                        <li style="color: black;">Optionally include an ID column (id, toi, or tid) for identification</li>
+                        <li style="color: black;">Use 0 or NaN for unknown/missing values</li>
+                    </ul>
+                </div>
+            </div>
+            """)
+            
+            with gr.Row():
+                with gr.Column(scale=1):
+                    gr.HTML("""
+                    <div style="background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); 
+                                padding: 20px; border-radius: 12px; margin-bottom: 20px; text-align: center;">
+                        <h3 style="color: white; margin: 0; font-size: 22px;">⚙️ Configuration</h3>
+                    </div>
+                    """)
+                    
+                    single_file = gr.File(
+                        label="📁 Upload Single Sample CSV",
+                        file_types=[".csv"],
+                        type="filepath"
+                    )
+                    
+                    single_model = gr.Dropdown(
+                        choices=['Ensemble', 'LightGBM', 'XGBoost', 'RandomForest', 'ExtraTrees'],
+                        value='Ensemble',
+                        label="🤖 Select Classification Model",
+                        info="Ensemble combines multiple models (Recommended)"
+                    )
+                    
+                    single_predict_btn = gr.Button("🚀 Classify Sample", variant="primary", size="lg")
+                    
+                    gr.HTML("""
+                    <div style="background: linear-gradient(to right, #fff9e6, #fffef0); 
+                                border-left: 5px solid #ff9800; padding: 20px; border-radius: 10px; margin-top: 20px;">
+                        <h4 style="color: black; margin: 0 0 10px 0;">⚡ Quick Requirements:</h4>
+                        <ul style="color: black; margin: 0; padding-left: 20px; line-height: 2;">
+                            <li style="color: black;"><strong style="color: black;">Exactly 2 rows</strong> (header + 1 data row)</li>
+                            <li style="color: black;">All <strong style="color: black;">52 feature columns</strong> required</li>
+                            <li style="color: black;">Optional ID column for tracking</li>
+                        </ul>
+                    </div>
+                    
+                    <div style="background: #e8f5e9; border-left: 5px solid #4caf50; 
+                                padding: 20px; border-radius: 10px; margin-top: 15px;">
+                        <h4 style="color: black; margin: 0 0 10px 0;">💡 Need Help?</h4>
+                        <p style="color: black; margin: 0 0 10px 0; font-size: 14px;">
+                            Download a properly formatted template CSV with all required columns:
+                        </p>
+                    """)
+                    
+                    single_template_btn = gr.Button("📥 Download Template CSV", variant="secondary")
+                    single_template_file = gr.File(label="Template File")
+                    
+                    gr.HTML("</div>")
+                    
+                    # Feature Requirements Section
+                    with gr.Accordion("📋 Required Feature Columns (52)", open=False):
+                        feature_table = """<div style='max-height: 500px; overflow-y: auto; padding: 10px;'>
+                        <table style='width: 100%; border-collapse: collapse; font-size: 13px;'>
+                        <thead style='position: sticky; top: 0; background: #667eea; color: white;'>
+                        <tr>
+                            <th style='padding: 10px; text-align: left; border: 1px solid #ddd;'>Column Name</th>
+                            <th style='padding: 10px; text-align: left; border: 1px solid #ddd;'>Parameter</th>
+                            <th style='padding: 10px; text-align: left; border: 1px solid #ddd;'>Description</th>
+                        </tr>
+                        </thead>
+                        <tbody>"""
+                        
+                        for i, feat in enumerate(BASE_FEATURES):
+                            label, desc = PARAM_INFO[feat]
+                            bg_color = '#f9f9f9' if i % 2 == 0 else 'white'
+                            feature_table += f"""
+                        <tr style='background: {bg_color};'>
+                            <td style='padding: 8px; border: 1px solid #ddd; font-family: monospace; color: #c7254e; background-color: #f9f2f4; font-weight: bold;'>{feat}</td>
+                            <td style='padding: 8px; border: 1px solid #ddd; font-weight: 600; color: #333;'>{label}</td>
+                            <td style='padding: 8px; border: 1px solid #ddd; color: #555;'>{desc}</td>
+                        </tr>"""
+                        
+                        feature_table += """</tbody></table></div>"""
+                        gr.HTML(feature_table)
+                
+                with gr.Column(scale=2):
+                    gr.HTML("""
+                    <div style="background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); 
+                                padding: 20px; border-radius: 12px; margin-bottom: 20px;">
+                        <h2 style="color: white; margin: 0; text-align: center; font-size: 28px;">
+                            📊 Classification Results
+                        </h2>
+                    </div>
+                    """)
+                    
+                    single_result_display = gr.HTML(label="Prediction Result")
+                    single_prob_display = gr.HTML(label="Class Probabilities")
+                    single_preview = gr.HTML(label="Input Summary")
+            
+            single_predict_btn.click(
+                fn=predict_single,
+                inputs=[single_file, single_model],
+                outputs=[single_result_display, single_prob_display, single_preview]
+            )
+            
+            single_template_btn.click(fn=get_template, outputs=single_template_file)
+        
+        # ===== BATCH TAB =====
+        with gr.Tab("📦 Batch Processing"):
+            gr.HTML("""
+            <div style="background: linear-gradient(to right, #fff3e6, #f5f5f5); padding: 25px; 
+                        border-radius: 15px; border-left: 5px solid #ff9800; margin-bottom: 25px;">
+                <h2 style="color: #e65100; margin: 0 0 10px 0;">Bulk CSV Processing</h2>
+                <p style="color: #424242; margin: 0; font-size: 16px; line-height: 1.6;">
+                    Upload a CSV file containing multiple samples. Your file must include all 52 required feature columns.
+                    Results will be provided in a simple ID-Label format ready for download.
+                </p>
+            </div>
+            """)
+            
+            # Feature Requirements Section for Batch
+            with gr.Accordion("📋 Required Feature Columns (52) - Click to View Details", open=False):
+                gr.HTML("""
+                <div style="background: linear-gradient(to right, #fff3e0, #fffbf0); padding: 20px; 
+                            border-radius: 10px; border-left: 5px solid #ff9800; margin-bottom: 15px;">
+                    <h3 style="color: #e65100; margin: 0 0 10px 0;">📌 CSV Format Requirements</h3>
+                    <ul style="color: #5d4037; line-height: 1.8; margin: 0;">
+                        <li>Your CSV must contain <strong>all 52 required feature columns</strong> listed in the table below</li>
+                        <li>Column names must <strong>exactly match</strong> the names shown (case-sensitive)</li>
+                        <li>Optionally include an ID column (<code>id</code>, <code>toi</code>, or <code>tid</code>) to track your samples</li>
+                        <li>Use <code>0</code> or leave empty for unknown/missing values (will be imputed automatically)</li>
+                        <li>Each row represents one exoplanet candidate to classify</li>
+                        <li>Can process multiple samples at once (no row limit)</li>
+                    </ul>
+                </div>
+                """)
+                
+                batch_feature_table = """<div style='max-height: 500px; overflow-y: auto; padding: 10px; background: white; border-radius: 8px; box-shadow: 0 2px 8px rgba(0,0,0,0.1);'>
+                <table style='width: 100%; border-collapse: collapse; font-size: 13px;'>
+                <thead style='position: sticky; top: 0; background: #ff9800; color: white; z-index: 10;'>
+                <tr>
+                    <th style='padding: 12px; text-align: left; border: 1px solid #ddd;'>Column Name</th>
+                    <th style='padding: 12px; text-align: left; border: 1px solid #ddd;'>Parameter</th>
+                    <th style='padding: 12px; text-align: left; border: 1px solid #ddd;'>Description</th>
+                </tr>
+                </thead>
+                <tbody>"""
+                
+                for i, feat in enumerate(BASE_FEATURES):
+                    label, desc = PARAM_INFO[feat]
+                    bg_color = '#fff3e0' if i % 2 == 0 else 'white'
+                    batch_feature_table += f"""
+                <tr style='background: {bg_color};'>
+                    <td style='padding: 8px; border: 1px solid #ddd; font-family: monospace; color: #c7254e; background-color: #f9f2f4; font-weight: bold;'>{feat}</td>
+                    <td style='padding: 8px; border: 1px solid #ddd; font-weight: 600; color: #333;'>{label}</td>
+                    <td style='padding: 8px; border: 1px solid #ddd; color: #555;'>{desc}</td>
+                </tr>"""
+                
+                batch_feature_table += """</tbody></table></div>"""
+                gr.HTML(batch_feature_table)
+            
+            with gr.Row():
+                with gr.Column(scale=1):
+                    batch_file = gr.File(label="📁 Upload CSV File", file_types=[".csv"])
+                    batch_model = gr.Dropdown(
+                        choices=['Ensemble', 'LightGBM', 'XGBoost', 'RandomForest', 'ExtraTrees'],
+                        value='Ensemble',
+                        label="🤖 Classification Model"
+                    )
+                    batch_btn = gr.Button("🚀 Process Batch", variant="primary", size="lg")
+                    
+                    gr.HTML("""
+                    <div style="background: #e8f5e9; border-left: 5px solid #4caf50; 
+                                padding: 20px; border-radius: 10px; margin-top: 20px;">
+                        <h4 style="color: #2e7d32; margin-top: 0;">💡 Download Template</h4>
+                        <p style="color: #1b5e20; margin-bottom: 15px;">
+                            Need a template? Click below to download a properly formatted CSV file with all required columns.
+                        </p>
+                    """)
+                    template_btn = gr.Button("📥 Download Template CSV", variant="secondary")
+                    template_output = gr.File(label="Template File")
+                    gr.HTML("</div>")
+                
+                with gr.Column(scale=2):
+                    batch_summary = gr.Markdown()
+                    batch_results = gr.Dataframe(label="Results Preview")
+                    batch_download = gr.File(label="📥 Download Results (ID, Label)")
+            
+            batch_btn.click(
+                fn=predict_batch,
+                inputs=[batch_file, batch_model],
+                outputs=[batch_summary, batch_results, batch_download]
+            )
+            
+            template_btn.click(fn=get_template, outputs=template_output)
+    
+    # Footer
+    gr.HTML("""
+    <div style="margin-top: 50px; padding: 30px; background: linear-gradient(135deg, #1a1a2e 0%, #16213e 100%); 
+                border-radius: 15px; text-align: center; box-shadow: 0 8px 20px rgba(0,0,0,0.4);">
+        <h3 style="color: #e0e0e0; margin: 0 0 15px 0;">🌟 Exoplanet Classification System</h3>
+        <p style="color: #b0b0b0; font-size: 15px; margin: 10px 0;">
+            Powered by: <strong style="color: #88c0d0;">LightGBM</strong> • <strong style="color: #81a1c1;">XGBoost</strong> • <strong style="color: #5e81ac;">Random Forest</strong> • <strong style="color: #8fbcbb;">Scikit-learn</strong>
+        </p>
+        <p style="color: #888; font-size: 13px; margin: 10px 0;">
+            Data: TESS Mission | Built with Gradio | Version 2025.10.04
+        </p>
+        <p style="color: #666; font-size: 12px; margin: 10px 0;">
+            ⚠️ For research and educational purposes. Validate results independently.
+        </p>
+    </div>
+    """)
+
+# ====================================================================
+# LAUNCH
+# ====================================================================
+
+if __name__ == "__main__":
+    print("\n" + "="*70)
+    print(" " * 20 + "LAUNCHING WEB APPLICATION")
+    print("="*70 + "\n")
+    
+    app.launch(
+        server_name="0.0.0.0",
+        server_port=7860,
+        show_error=True,
+        share=True
+    )

mutiresult/classification_report_Ensemble.txt

@@ -0,0 +1,15 @@
+Classification Report: Ensemble
+======================================================================
+
+              precision    recall  f1-score   support
+
+         APC     0.5000    0.2105    0.2963        19
+          CP     0.3500    0.2121    0.2642        33
+          FA     0.0000    0.0000    0.0000         1
+          FP     0.3333    0.1400    0.1972        50
+          KP     0.7143    0.5556    0.6250        54
+          PC     0.7638    0.9157    0.8329       332
+
+    accuracy                         0.7198       489
+   macro avg     0.4436    0.3390    0.3693       489
+weighted avg     0.6746    0.7198    0.6840       489

mutiresult/classification_report_ExtraTrees.txt

@@ -0,0 +1,15 @@
+Classification Report: ExtraTrees
+======================================================================
+
+              precision    recall  f1-score   support
+
+         APC     0.5000    0.0526    0.0952        19
+          CP     0.2727    0.0909    0.1364        33
+          FA     0.0000    0.0000    0.0000         1
+          FP     0.2500    0.0400    0.0690        50
+          KP     0.8462    0.4074    0.5500        54
+          PC     0.7195    0.9578    0.8217       332
+
+    accuracy                         0.7076       489
+   macro avg     0.4314    0.2581    0.2787       489
+weighted avg     0.6453    0.7076    0.6386       489

mutiresult/classification_report_LightGBM.txt

@@ -0,0 +1,15 @@
+Classification Report: LightGBM
+======================================================================
+
+              precision    recall  f1-score   support
+
+         APC     0.4167    0.2632    0.3226        19
+          CP     0.3214    0.2727    0.2951        33
+          FA     0.0000    0.0000    0.0000         1
+          FP     0.4000    0.2400    0.3000        50
+          KP     0.6000    0.6111    0.6055        54
+          PC     0.7995    0.8765    0.8362       332
+
+    accuracy                         0.7157       489
+   macro avg     0.4229    0.3773    0.3932       489
+weighted avg     0.6878    0.7157    0.6977       489

mutiresult/classification_report_LogisticRegression.txt

@@ -0,0 +1,15 @@
+Classification Report: LogisticRegression
+======================================================================
+
+              precision    recall  f1-score   support
+
+         APC     0.1739    0.6316    0.2727        19
+          CP     0.2278    0.5455    0.3214        33
+          FA     0.0000    0.0000    0.0000         1
+          FP     0.4333    0.5200    0.4727        50
+          KP     0.3626    0.6111    0.4552        54
+          PC     0.9198    0.5181    0.6628       332
+
+    accuracy                         0.5337       489
+   macro avg     0.3529    0.4710    0.3641       489
+weighted avg     0.7310    0.5337    0.5809       489

mutiresult/classification_report_RandomForest.txt

@@ -0,0 +1,15 @@
+Classification Report: RandomForest
+======================================================================
+
+              precision    recall  f1-score   support
+
+         APC     0.0000    0.0000    0.0000        19
+          CP     0.2222    0.0606    0.0952        33
+          FA     0.0000    0.0000    0.0000         1
+          FP     0.1429    0.0200    0.0351        50
+          KP     0.8462    0.4074    0.5500        54
+          PC     0.7152    0.9608    0.8201       332
+
+    accuracy                         0.7035       489
+   macro avg     0.3211    0.2415    0.2501       489
+weighted avg     0.6087    0.7035    0.6275       489

mutiresult/classification_report_XGBoost.txt

@@ -0,0 +1,15 @@
+Classification Report: XGBoost
+======================================================================
+
+              precision    recall  f1-score   support
+
+         APC     0.5000    0.2105    0.2963        19
+          CP     0.3478    0.2424    0.2857        33
+          FA     0.0000    0.0000    0.0000         1
+          FP     0.3333    0.1400    0.1972        50
+          KP     0.7143    0.5556    0.6250        54
+          PC     0.7696    0.9157    0.8363       332
+
+    accuracy                         0.7219       489
+   macro avg     0.4442    0.3440    0.3734       489
+weighted avg     0.6784    0.7219    0.6878       489

mutiresult/feature_info.json

@@ -0,0 +1,126 @@
+{
+    "all_features": [
+        "ra",
+        "dec",
+        "st_pmra",
+        "st_pmraerr1",
+        "st_pmraerr2",
+        "st_pmralim",
+        "st_pmdec",
+        "st_pmdecerr1",
+        "st_pmdecerr2",
+        "st_pmdeclim",
+        "pl_tranmid",
+        "pl_tranmiderr1",
+        "pl_tranmiderr2",
+        "pl_tranmidlim",
+        "pl_orbper",
+        "pl_orbpererr1",
+        "pl_orbpererr2",
+        "pl_orbperlim",
+        "pl_trandurh",
+        "pl_trandurherr1",
+        "pl_trandurherr2",
+        "pl_trandurhlim",
+        "pl_trandep",
+        "pl_trandeperr1",
+        "pl_trandeperr2",
+        "pl_trandeplim",
+        "pl_rade",
+        "pl_radeerr1",
+        "pl_radeerr2",
+        "pl_radelim",
+        "pl_insol",
+        "pl_insolerr1",
+        "pl_insolerr2",
+        "pl_insollim",
+        "pl_eqt",
+        "pl_eqterr1",
+        "pl_eqterr2",
+        "pl_eqtlim",
+        "st_tmag",
+        "st_tmagerr1",
+        "st_tmagerr2",
+        "st_tmaglim",
+        "st_dist",
+        "st_disterr1",
+        "st_disterr2",
+        "st_distlim",
+        "st_teff",
+        "st_tefferr1",
+        "st_tefferr2",
+        "st_tefflim",
+        "st_logg",
+        "st_loggerr1",
+        "st_loggerr2",
+        "st_logglim",
+        "st_rad",
+        "st_raderr1",
+        "st_raderr2",
+        "st_radlim"
+    ],
+    "selected_features": [
+        "ra",
+        "dec",
+        "st_pmra",
+        "st_pmraerr1",
+        "st_pmraerr2",
+        "st_pmralim",
+        "st_pmdec",
+        "st_pmdecerr1",
+        "st_pmdecerr2",
+        "st_pmdeclim",
+        "pl_tranmid",
+        "pl_tranmiderr1",
+        "pl_tranmiderr2",
+        "pl_tranmidlim",
+        "pl_orbper",
+        "pl_orbpererr1",
+        "pl_orbpererr2",
+        "pl_trandurh",
+        "pl_trandurherr1",
+        "pl_trandurherr2",
+        "pl_trandep",
+        "pl_trandeperr1",
+        "pl_trandeperr2",
+        "pl_rade",
+        "pl_radeerr1",
+        "pl_radeerr2",
+        "pl_radelim",
+        "pl_insol",
+        "pl_eqt",
+        "st_tmag",
+        "st_tmagerr1",
+        "st_dist",
+        "st_disterr1",
+        "st_disterr2",
+        "st_tefferr1",
+        "st_tefferr2",
+        "st_tefflim",
+        "st_logg",
+        "st_loggerr1",
+        "st_loggerr2",
+        "st_rad",
+        "st_raderr1",
+        "st_raderr2",
+        "st_radlim",
+        "luminosity_proxy",
+        "planet_star_radius_ratio",
+        "transit_depth_proxy",
+        "log_distance",
+        "inv_distance",
+        "temp_ratio",
+        "log_insol",
+        "sqrt_insol",
+        "log_orbper",
+        "st_pmra_error_ratio",
+        "st_pmdec_error_ratio",
+        "pl_rade_error_ratio",
+        "feature_mean",
+        "feature_std",
+        "feature_max",
+        "feature_min"
+    ],
+    "n_features": 58,
+    "n_selected": 60
+}

mutiresult/feature_scores.csv

@@ -0,0 +1,69 @@
+feature,score
+pl_tranmid,0.10058388860470635
+pl_trandurherr2,0.0943360041537582
+pl_trandurherr1,0.09260907503584814
+st_tmag,0.0773120734401902
+pl_rade_error_ratio,0.0746345642044055
+pl_tranmiderr1,0.07072746195404522
+pl_tranmiderr2,0.06939452988563999
+pl_trandeperr2,0.06708691436040959
+pl_trandeperr1,0.06708616815245194
+st_rad,0.0653051637978963
+st_loggerr1,0.060289975854914246
+log_orbper,0.05710053100824575
+pl_orbper,0.05673082307809074
+st_loggerr2,0.055891289285257884
+st_tefferr1,0.054497754817492305
+st_dist,0.054088027146716566
+inv_distance,0.053659444012782664
+log_distance,0.05353421359003718
+st_tefferr2,0.05252489140575456
+pl_trandurh,0.050686071123912546
+st_pmraerr2,0.047570679656092185
+dec,0.04553914467762987
+st_raderr2,0.038896264430238414
+st_pmdecerr2,0.03888845727031187
+st_disterr1,0.03736688469809901
+st_disterr2,0.03631174509371071
+pl_orbpererr1,0.03599140473565621
+pl_rade,0.0359609738241089
+pl_insol,0.034403756845682976
+sqrt_insol,0.03417849794237493
+log_insol,0.03405330470849344
+st_raderr1,0.03337667004872613
+pl_radeerr1,0.029370008341174758
+pl_radeerr2,0.029187264207060748
+st_pmdecerr1,0.0290700400755326
+temp_ratio,0.028656777472438222
+pl_orbpererr2,0.02803107120731263
+st_pmraerr1,0.02736478939231235
+st_logg,0.027120721911320533
+feature_min,0.027058757264905253
+pl_trandep,0.025588020080648866
+pl_eqt,0.02416596292758255
+planet_star_radius_ratio,0.018561314834198983
+transit_depth_proxy,0.018370615691842174
+feature_max,0.01766602864717748
+feature_std,0.017552613001937445
+feature_mean,0.01718836978668059
+st_pmdec,0.014548337924330523
+ra,0.014342613726746167
+pl_radelim,0.014038127813360113
+st_pmra,0.010297755012506737
+st_pmra_error_ratio,0.008262326729144398
+luminosity_proxy,0.0078110578935157715
+st_pmdeclim,0.006311307886967432
+st_tefflim,0.005229401126498834
+pl_tranmidlim,0.004476832142852993
+st_tmagerr1,0.0036663497865565553
+st_pmralim,0.003033599214105731
+st_radlim,0.0030230088374789688
+st_pmdec_error_ratio,0.0022816723367815595
+pl_trandurhlim,0.0007805835305294551
+st_tmaglim,0.00019178201976965958
+pl_orbperlim,0.0
+pl_trandeplim,0.0
+st_tmagerr2,0.0
+st_logglim,0.0
+st_distlim,0.0
+st_teff,0.0

mutiresult/imputer.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f23451bb545bd698bf5cb04f912f0b4e671e9d81bb9b0d652c6fd1ac264e67ed
+size 3606665

mutiresult/label_encoder.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0034712499e29d1d1afdeac2af29869ef79da2b941af6d7807f6edbde92cd20d
+size 504

mutiresult/metrics_summary.json

@@ -0,0 +1,62 @@
+{
+    "LightGBM": {
+        "Accuracy": 0.7157464212678937,
+        "Precision": 0.6878147074466093,
+        "Recall": 0.7157464212678937,
+        "F1-Score": 0.6977191708337254,
+        "MCC": 0.40092736912599086,
+        "Cohen_Kappa": 0.3966231990838963,
+        "Log_Loss": 0.8667761106577778,
+        "AUC_OVR": 0.8349458443198322
+    },
+    "XGBoost": {
+        "Accuracy": 0.721881390593047,
+        "Precision": 0.6783850309985548,
+        "Recall": 0.721881390593047,
+        "F1-Score": 0.6877780716424755,
+        "MCC": 0.37628533150933247,
+        "Cohen_Kappa": 0.35900994679620624,
+        "Log_Loss": 0.8242094430417346,
+        "AUC_OVR": 0.8421016022429259
+    },
+    "RandomForest": {
+        "Accuracy": 0.7034764826175869,
+        "Precision": 0.6086510870686501,
+        "Recall": 0.7034764826175869,
+        "F1-Score": 0.6275139315266048,
+        "MCC": 0.2606656979570372,
+        "Cohen_Kappa": 0.20312658042909004,
+        "Log_Loss": 0.7921825120152793,
+        "AUC_OVR": 0.8300185572722183
+    },
+    "ExtraTrees": {
+        "Accuracy": 0.7075664621676891,
+        "Precision": 0.6453006875237117,
+        "Recall": 0.7075664621676891,
+        "F1-Score": 0.6385766896155018,
+        "MCC": 0.2798313236799552,
+        "Cohen_Kappa": 0.2245153706250277,
+        "Log_Loss": 0.7800578509848219,
+        "AUC_OVR": 0.8283101428969134
+    },
+    "LogisticRegression": {
+        "Accuracy": 0.5337423312883436,
+        "Precision": 0.7309640150800872,
+        "Recall": 0.5337423312883436,
+        "F1-Score": 0.5808969016846894,
+        "MCC": 0.361136062774439,
+        "Cohen_Kappa": 0.32511712933257464,
+        "Log_Loss": 1.2702846778722883,
+        "AUC_OVR": 0.794391056021819
+    },
+    "Ensemble": {
+        "Accuracy": 0.7198364008179959,
+        "Precision": 0.6745931184797294,
+        "Recall": 0.7198364008179959,
+        "F1-Score": 0.6839894359599863,
+        "MCC": 0.3674686647252085,
+        "Cohen_Kappa": 0.3486655971999416,
+        "Log_Loss": 0.7483648859389048,
+        "AUC_OVR": 0.8429211659418736
+    }
+}

mutiresult/model_extratrees.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:790de4aa3a17d30ac083800263bc9d85b1e6273aba763304f336890e76b70f23
+size 84380265

mutiresult/model_lightgbm.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e02771e8251b13456b2d3dc9d089d9a344de9bfae17c6e1137fc2935dcfd8e47
+size 5715044

mutiresult/model_logisticregression.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ca452867d013f5f0f67e74c041eac577b1dd12a79138caa62375acf06621b9e
+size 4935

mutiresult/model_randomforest.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80958dad24deb019fcbb3e831a123d8767b64ce926dfa19919e675c50c558680
+size 50616470

mutiresult/model_xgboost.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:047dd9e61e19370d607592f79b8730ecfaf7096f55c3295c945d806305ac5520
+size 7511049

mutiresult/scaler.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0eca4f5252a5351a74a7c90f7393c3fe71385c235b2ccb8b0362d70e661ae8a
+size 2807

requirements.txt

@@ -0,0 +1,7 @@
+gradio==4.44.0
+pandas==2.2.3
+numpy==1.26.4
+scikit-learn==1.5.2
+joblib==1.4.2
+lightgbm==4.5.0
+xgboost==2.1.1