Delete app.py

app.py

@@ -1,80 +0,0 @@
-import gradio as gr
-import pandas as pd
-import numpy as np
-import xgboost as xgb
-from pybaseball import pitching_stats, cache
-import matplotlib.pyplot as plt
-
-cache.enable()
-
-class DecadalEngine:
-    def __init__(self):
-        raw_data = pitching_stats(2015, 2025, ind=1)
-        if 'Ball%' not in raw_data.columns: raw_data['Ball%'] = 0.35
-        if 'SIERA' not in raw_data.columns: raw_data['SIERA'] = raw_data['FIP']
-        self.features = ['SwStr%', 'Ball%', 'SIERA', 'GB%', 'FIP', 'Age', 'IP']
-        df = raw_data[self.features + ['SO', 'Name', 'Season']].dropna()
-        dtrain = xgb.DMatrix(df[self.features], label=df['SO'])
-        self.model = xgb.train({'objective': 'reg:squarederror', 'max_depth': 5}, dtrain)
-        self.latest_stats = df[df['Season'] == 2025]
-
-engine = DecadalEngine()
-
-def run_visual_sim(p1, p2, conf_str):
-    z_map = {"90%": 1.645, "95%": 1.96, "99% (Ultra-Precision)": 2.576}
-    z_val = z_map[conf_str]
-    
-    names = [n for n in [p1, p2] if n]
-    projections, lowers, uppers = [], [], []
-    
-    for name in names:
-        player_data = engine.latest_stats[engine.latest_stats['Name'] == name]
-        input_row = player_data[engine.features].copy().iloc[0:1]
-        input_row['Age'] += 1 
-        
-        mu = float(engine.model.predict(xgb.DMatrix(input_row))[0])
-        std_error = np.sqrt(mu) * (0.6 if player_data['SIERA'].values[0] < 3.4 else 0.9)
-        
-        projections.append(mu)
-        lowers.append(mu - (z_val * std_error))
-        uppers.append(mu + (z_val * std_error))
-
-    # --- Create Horizontal Bar Chart ---
-    fig, ax = plt.subplots(figsize=(10, 4), facecolor='#f8f9fa')
-    y_pos = np.arange(len(names))
-    # Error bar needs relative lengths: (mu - low) and (high - mu)
-    xerr = [[p - l for p, l in zip(projections, lowers)], 
-            [u - p for u, p in zip(uppers, projections)]]
-    
-    bars = ax.barh(y_pos, projections, xerr=xerr, capsize=10, 
-                   color=['#003566', '#d62828'], alpha=0.7, height=0.6,
-                   error_kw={'ecolor': 'black', 'lw': 2, 'capthick': 2})
-    
-    ax.set_yticks(y_pos)
-    ax.set_yticklabels(names, fontweight='bold', fontsize=12)
-    ax.set_xlabel("Projected 2026 Strikeouts", fontsize=10)
-    ax.set_title(f"2026 Strikeout Projections with {conf_str} Confidence Intervals", fontweight='bold')
-    
-    # Label the bars with the exact projection value
-    for i, v in enumerate(projections):
-        ax.text(v + 3, i, f"{int(v)} K", va='center', fontweight='bold', color='black')
-
-    plt.tight_layout()
-    
-    summary = "\n\n".join([f"**{names[i]}**: {int(projections[i])} Ks (Range: {int(lowers[i])}–{int(uppers[i])})" for i in range(len(names))])
-    return fig, summary
-
-with gr.Blocks(theme=gr.themes.Soft()) as demo:
-    gr.Markdown("# 📊 MLB 2026 Visual Prediction Engine")
-    with gr.Row():
-        with gr.Column(scale=1):
-            p1 = gr.Dropdown(choices=sorted(engine.latest_stats['Name'].tolist()), label="Pitcher 1", value="Jesus Luzardo")
-            p2 = gr.Dropdown(choices=sorted(engine.latest_stats['Name'].tolist()), label="Pitcher 2", value="Andre Pallante")
-            conf = gr.Radio(["90%", "95%", "99% (Ultra-Precision)"], label="Confidence Level", value="95%")
-            btn = gr.Button("🚀 Generate Visual Report", variant="primary")
-        with gr.Column(scale=2):
-            plot, txt = gr.Plot(), gr.Markdown()
-            
-    btn.click(run_visual_sim, inputs=[p1, p2, conf], outputs=[plot, txt])
-
-demo.launch()