Update app.py

app.py

@@ -1,164 +1,278 @@
-# app.py
-import gradio as gr
-import pandas as pd
+# app.py  –  all-in-one  Financial-OS v6.0 for Hugging Face Spaces
+from __future__ import annotations
+import io
+import logging
+import warnings
+from dataclasses import dataclass
+from typing import Dict, Protocol, Tuple
+import datetime as dt
+
 import numpy as np
-from pathlib import Path
+import pandas as pd
+import torch
+import torch.nn as nn
+from sklearn.decomposition import PCA
+from sklearn.preprocessing import StandardScaler
+from sklearn.model_selection import TimeSeriesSplit
+import xgboost as xgb
+import lightgbm as lgb
+import gradio as gr
 from reportlab.lib.pagesizes import A4
 from reportlab.pdfgen import canvas
 from reportlab.pdfbase.ttfonts import TTFont
 from reportlab.pdfbase import pdfmetrics
 import textwrap
-import io
-import datetime as dt
+import requests
+import os
+
+warnings.filterwarnings("ignore")
+
+# --------------- 1.  Prometheus metrics (optional) ---------------
+# prometheus_client จะถูกติดตั้งตาม requirements.txt
+try:
+    from prometheus_client import Counter, Histogram, Gauge
+    PRED_COUNT = Counter("fos_pred_total", "Total predictions")
+    LATENCY  = Histogram("fos_latency_sec", "Latency")
+    CORR_G   = Gauge("fos_corr", "Last correlation")
+    SHARPE_G = Gauge("fos_sharpe", "Last Sharpe")
+    METRICS_ON = True
+except Exception:
+    METRICS_ON = False
+
+# --------------- 2.  Protocol & Value Objects ---------------
+class MLModel(Protocol):
+    def fit(self, X: np.ndarray, y: np.ndarray) -> None: ...
+    def predict(self, X: np.ndarray) -> np.ndarray: ...
 
-# โหลด engine ที่ refactor ไว้แล้ว
-from financial_os_v6_hf import FinancialOSv6, show_banner
-
-# ลงทะเบียนฟอนต์ภาษาไทย (ใช้ Sarabun ของ Google)
-FONT_URL = "https://github.com/google/fonts/raw/main/ofl/sarabun/Sarabun-Regular.ttf"
-font_path = Path("Sarabun-Regular.ttf")
-if not font_path.exists():
-    import requests, os
-    with open(font_path, "wb") as f:
-        f.write(requests.get(FONT_URL).content)
-pdfmetrics.registerFont(TTFont("Sarabun", str(font_path)))
-
-# สร้าง PDF รายงาน
-def create_pdf_report(metrics, bvi):
-    buffer = io.BytesIO()
-    c = canvas.Canvas(buffer, pagesize=A4)
-    width, height = A4
-    c.setFont("Sarabun", 14)
-
-    def draw_thai(text, x, y, max_width=110):
-        lines = textwrap.wrap(text, width=max_width)
-        for line in lines:
-            c.drawString(x, y, line)
-            y -= 20
+@dataclass
+class Metrics:
+    correlation: float
+    r_squared: float
+    mae: float
+    sharpe: float
+
+@dataclass
+class BVIResult:
+    score: float
+    components: Dict[str, float]
+    risk_level: str
+    recommendation: str
+    projected_roi: float
+    confidence: Tuple[float, float]
+    world_ranking: str
+
+# --------------- 3.  Engines ---------------
+class XGBEngine:
+    def __init__(self) -> None:
+        self.model: MLModel = xgb.XGBRegressor(
+            n_estimators=1000, learning_rate=0.01, max_depth=7,
+            subsample=0.8, colsample_bytree=0.8, n_jobs=-1, verbosity=0)
+    def fit(self, X: np.ndarray, y: np.ndarray) -> None: self.model.fit(X, y)
+    def predict(self, X: np.ndarray) -> np.ndarray: return self.model.predict(X)
+
+class FeatureExtractor:
+    def __init__(self) -> None:
+        self.pca = PCA(n_components=3)
+        self.scaler = StandardScaler()
+    def transform(self, df: pd.DataFrame) -> pd.DataFrame:
+        df = df.copy()
+        fin = ["revenue_ttm", "burn_rate_monthly", "cash_reserves"]
+        sc = self.scaler.fit_transform(df[fin])
+        pc = self.pca.fit_transform(sc)
+        df["financial_health_score"] = pc @ np.array([0.6, 0.3, 0.1])
+        df["growth_momentum"] = df["revenue_growth_3y"]*0.7 + np.log1p(df["revenue_ttm"])*0.3
+        df["team_execution_power"] = (df["team_cohesion_score"]**2 *0.4 +
+                                      df["execution_velocity"]**1.5 *0.4 +
+                                      df["founder_commitment_score"]**2 *0.2)
+        df["market_timing_advantage"] = np.expm1(df["market_timing_score"] * df["sector_heat_score"])
+        df["competitive_moat"] = 1 - df["competitive_landscape_score"]
+        df["hidden_gem_boost"] = np.where((df["revenue_growth_3y"]>0.6)&(df["team_execution_power"]>0.7), 1.3, 1.0)
+        df["risk_adjusted_growth"] = df["revenue_growth_3y"] / (1 + np.abs(df["profit_margin"]))
+        df["cash_efficiency"] = np.where(df["burn_rate_monthly"]>0,
+                                         df["revenue_ttm"]/(df["burn_rate_monthly"]*12), 1.0)
+        df["pmf_momentum"] = df["product_market_fit"] * df["growth_momentum"]
+        df["quantum_risk_adjustment"] = 1 - (0.3*(1-df["regulatory_environment_score"]) +
+                                             0.2*df["competitive_landscape_score"] +
+                                             0.1*(1-df["financial_health_score"]))
+        df["growth_team_interaction"] = df["revenue_growth_3y"] * df["team_execution_power"]
+        df["market_financial_interaction"] = df["market_timing_advantage"] * df["financial_health_score"]
+        df["execution_moat_interaction"] = df["execution_velocity"] * df["competitive_moat"]
+        return df
+
+class BVIv6Calc:
+    WG = {"rev":0.45, "fin":0.25, "team":0.15, "mkt":0.10, "gov":0.05}
+    BOOST = 1.3; MAX_INT = 0.3
+    def calc(self, f: pd.Series) -> BVIResult:
+        g = f["revenue_growth_3y"]*0.7 + f["growth_momentum"]*0.3
+        fi = f["financial_health_score"]*0.6 + f["cash_efficiency"]*0.3 + max(f["profit_margin"],0)*0.1
+        t = f["team_execution_power"]*0.5 + f["execution_velocity"]*0.3 + f["founder_commitment_score"]*0.2
+        m = f["market_timing_advantage"]*0.4 + f["sector_heat_score"]*0.3 + f["competitive_moat"]*0.2 + f["regulatory_environment_score"]*0.1
+        gov = f["transparency_score"]*0.4 + f["compliance_score"]*0.3 + f["board_independence"]*0.3
+        base = (g*self.WG["rev"] + fi*self.WG["fin"] + t*self.WG["team"] +
+                m*self.WG["mkt"] + gov*self.WG["gov"])*100
+        if f["hidden_gem_boost"]>1.0: base *= self.BOOST
+        base *= f["quantum_risk_adjustment"]
+        inter = (f["growth_team_interaction"]*0.2 + f["market_financial_interaction"]*0.15 +
+                 f["execution_moat_interaction"]*0.1)
+        final = base * (1 + min(inter, self.MAX_INT))
+        risk, rec = self._risk(final)
+        roi, conf = self._roi(final, f)
+        rank = self._rank(final)
+        return BVIResult(score=round(final,1),
+                         components={"growth":round(g*100,1),"financial":round(fi*100,1),
+                                     "team":round(t*100,1),"market":round(m*100,1),
+                                     "governance":round(gov*100,1)},
+                         risk_level=risk, recommendation=rec,
+                         projected_roi=round(roi,2), confidence=conf, world_ranking=rank)
+    def _risk(self,s:float)->Tuple[str,str]:
+        if s>=85: return "🏆 WORLD-CLASS","Immediate investment"
+        if s>=75: return "🌟 EXCELLENT","Strong investment"
+        if s>=65: return "✅ GOOD","Recommended"
+        if s>=55: return "⚠️ MODERATE","Monitor closely"
+        if s>=45: return "🔴 HIGH RISK","Pilot only"
+        return "⛔ AVOID","Not recommended"
+    def _roi(self,s:float,f:pd.Series)->Tuple[float,Tuple[float,float]]:
+        base = s*0.04
+        g_boost = min(f["revenue_growth_3y"]/2,1.0)
+        t_boost = f["team_execution_power"]; m_boost = f["market_timing_advantage"]
+        roi = base*(1+g_boost*0.3)*(1+t_boost*0.2)*(1+m_boost*0.1)
+        roi = min(roi,20.0)
+        conf = (roi*0.85,roi*1.15) if s>=80 else ((roi*0.8,roi*1.2) if s>=65 else (roi*0.7,roi*1.3))
+        return roi, conf
+    def _rank(self,s:float)->str:
+        if s>=90: return "Top 1% Globally 🏆"
+        if s>=80: return "Top 5% Globally 🌟"
+        if s>=70: return "Top 15% Globally ✅"
+        if s>=60: return "Top 30% Globally 📊"
+        if s>=50: return "Top 50% Globally 📈"
+        return "Below Global Average 📉"
+
+class Backtester:
+    BENCH={"renaissance":{"sharpe":3.3},"biocatch":{"auc":0.97},"vantagescore":{"mae":0.05}}
+    def __init__(self,n_splits:int=5):self.tscv=TimeSeriesSplit(n_splits=n_splits)
+    def run(self,X:np.ndarray,y:np.ndarray,model:MLModel)->Metrics:
+        cors,r2s,maes,sharpes=[],[],[],[]
+        for tr,te in self.tscv.split(X):
+            model.fit(X[tr],y[tr]); pred=model.predict(X[te])
+            cors.append(np.corrcoef(y[te],pred)[0,1])
+            ss_res=np.sum((y[te]-pred)**2); ss_tot=np.sum((y[te]-y[te].mean())**2)
+            r2s.append(1-ss_res/ss_tot); maes.append(np.mean(np.abs(y[te]-pred)))
+            signal=np.where(pred>np.percentile(pred,70),1,-1); rets=y[te]*signal; rets=rets[rets!=0]
+            sharpe=(np.mean(rets)/np.std(rets)*np.sqrt(252)) if len(rets)>1 else 0.0
+            sharpes.append(sharpe)
+        return Metrics(correlation=float(np.mean(cors)),r_squared=float(np.mean(r2s)),
+                       mae=float(np.mean(maes)),sharpe=float(np.mean(sharpes)))
+
+# --------------- 4.  Orchestrator ---------------
+class FinancialOSv6:
+    def __init__(self)->None:
+        self.model= XGBEngine()
+        self.extractor= FeatureExtractor()
+        self.calc= BVIv6Calc()
+        self.backtest= Backtester()
+    def evaluate(self,df:pd.DataFrame,target_col:str="actual_roi")->Dict:
+        if METRICS_ON: PRED_COUNT.inc()
+        with LATENCY.time() if METRICS_ON else nullcontext():
+            df=self.extractor.transform(df)
+            X=df.drop(columns=[target_col]).values
+            y=df[target_col].values
+            metrics=self.backtest.run(X,y,self.model)
+            if METRICS_ON:
+                CORR_G.set(metrics.correlation); SHARPE_G.set(metrics.sharpe)
+            bvi=self.calc.calc(df.iloc[0])
+            return {"metrics":metrics,"bvi_sample":bvi}
+
+# --------------- 5.  PDF generator ---------------
+def create_pdf(metrics:Metrics,bvi:BVIResult)->bytes:
+    buffer=io.BytesIO(); c=canvas.Canvas(buffer,pagesize=A4)
+    # font
+    url="https://github.com/google/fonts/raw/main/ofl/sarabun/Sarabun-Regular.ttf"
+    font_name="Sarabun"
+    if not Path("Sarabun.ttf").exists():
+        open("Sarabun.ttf","wb").write(requests.get(url).content)
+    pdfmetrics.registerFont(TTFont(font_name,"Sarabun.ttf"))
+    c.setFont(font_name,14)
+    def draw(txt,x,y,max_w=110):
+        for line in textwrap.wrap(txt,width=max_w): c.drawString(x,y,line); y-=20
         return y
+    y=750
+    y=draw("🏦 Financial-OS v6.0 – รายงานผลการประเมิน",50,y)
+    y=draw(f"📅 วันที่ {dt.datetime.now():%d/%m/%Y %H:%M:%S}",50,y)
+    y=draw(f"📊 ค่าสหสัมพันธ์: {metrics.correlation:.3f}",50,y)
+    y=draw(f"📈 Sharpe Ratio: {metrics.sharpe:.2f}",50,y)
+    y=draw(f"💰 Projected ROI: {bvi.projected_roi:.2f}x",50,y)
+    y=draw(f"🎯 ระดับความเสี่ยง: {bvi.risk_level}",50,y)
+    y=draw(f"🌍 อันดับโลก: {bvi.world_ranking}",50,y)
+    c.save(); buffer.seek(0); return buffer.read()
+
+# --------------- 6.  Gradio UI ---------------
+from contextlib import nullcontext   # ใช้กับ prometheus optional
 
-    # header
-    c.drawString(50, height - 50, "🏦 Financial-OS v6.0 – รายงานผลการประเมิน")
-    c.drawString(50, height - 75, f"วันที่ {dt.datetime.now():%d/%m/%Y %H:%M:%S}")
-
-    y = height - 110
-    y = draw_thai(f"📊 ค่าสหสัมพันธ์ (Correlation): {metrics.correlation:.3f}", 50, y)
-    y = draw_thai(f"📈 Sharpe Ratio: {metrics.sharpe:.2f}", 50, y)
-    y = draw_thai(f"💰 ผลตอบแทนคาดการณ์: {bvi.projected_roi:.2f}x", 50, y)
-    y = draw_thai(f"🎯 ระดับความเสี่ยง: {bvi.risk_level}", 50, y)
-    y = draw_thai(f"🌍 อันดับโลก: {bvi.world_ranking}", 50, y)
-
-    c.save()
-    buffer.seek(0)
-    return buffer
-
-# สร้างตัวอย่างข้อมูลสุ่ม (เหมือนเดิม)
-def dummy_data() -> pd.DataFrame:
-    rng = np.random.default_rng(42)
-    n = 1
-    return pd.DataFrame({
-        "revenue_ttm": rng.lognormal(14, 1.5, n),
-        "revenue_growth_3y": rng.beta(2, 3, n) * 5,
-        "burn_rate_monthly": rng.exponential(50_000, n),
-        "cash_reserves": rng.lognormal(15, 1, n),
-        "profit_margin": rng.normal(0.1, 0.2, n),
-        "team_cohesion_score": rng.beta(4, 1.5, n),
-        "execution_velocity": rng.beta(3, 2, n),
-        "founder_commitment_score": rng.beta(5, 1, n),
-        "market_timing_score": rng.beta(3, 3, n),
-        "sector_heat_score": rng.beta(2, 2, n),
-        "competitive_landscape_score": rng.beta(2, 2, n),
-        "regulatory_environment_score": rng.beta(3, 1, n),
-        "product_market_fit": rng.beta(2, 2, n),
-        "transparency_score": rng.beta(9, 1, n),
-        "compliance_score": rng.beta(9, 1, n),
-        "board_independence": rng.beta(8, 2, n),
-        "actual_roi": rng.lognormal(1.1, 0.5, n),
-    })
-
-# สร้าง Gradio interface
 def evaluate(
-    revenue_ttm,
-    revenue_growth_3y,
-    burn_rate_monthly,
-    cash_reserves,
-    profit_margin,
-    team_cohesion_score,
-    execution_velocity,
-    founder_commitment_score,
-    market_timing_score,
-    sector_heat_score,
-    competitive_landscape_score,
-    regulatory_environment_score,
-    product_market_fit,
-    transparency_score,
-    compliance_score,
-    board_independence,
-):
-    df = pd.DataFrame(locals(), index=[0])
-    fos = FinancialOSv6()
-    result = fos.evaluate(df)
-    m = result["metrics"]
-    b = result["bvi_sample"]
-    pdf_buffer = create_pdf_report(m, b)
-    return (
-        f"{m.correlation:.3f}",
-        f"{m.sharpe:.2f}",
-        f"{b.score}",
-        b.risk_level,
-        b.world_ranking,
-        f"{b.projected_roi:.2f}x",
-        pdf_buffer,
-    )
-
-# สร้าง UI
-with gr.Blocks(title="Financial-OS v6.0 – HF GPU", theme=gr.themes.Soft()) as demo:
-    gr.Markdown("# 🏦 Financial-OS v6.0 – Global Leader AI Banking Platform")
-    gr.Markdown("🎯 ระบบประเมินมูลค่า Startup ด้วย AI | Sharpe > 25 | Correlation > 0.75")
+    revenue_ttm, revenue_growth_3y, burn_rate_monthly, cash_reserves, profit_margin,
+    team_cohesion_score, execution_velocity, founder_commitment_score, market_timing_score,
+    sector_heat_score, competitive_landscape_score, regulatory_environment_score,
+    product_market_fit, transparency_score, compliance_score, board_independence):
+    df=pd.DataFrame(locals(),index=[0])
+    res=FinancialOSv6().evaluate(df)
+    m=res["metrics"]; b=res["bvi_sample"]
+    pdf_bytes=create_pdf(m,b)
+    return (f"{m.correlation:.3f}", f"{m.sharpe:.2f}", f"{b.score}",
+            b.risk_level, b.world_ranking, f"{b.projected_roi:.2f}x", pdf_bytes)
 
+with gr.Blocks(title="Financial-OS v6.0 – HF GPU",theme=gr.themes.Soft()) as demo:
+    gr.Markdown("# 🏦 Financial-OS v6.0 – Global Leader AI Banking Platform")
+    gr.Markdown("🎯 ประเมินมูลค่า Startup ด้วย AI | Sharpe > 25 | Correlation > 0.75")
     with gr.Row():
         with gr.Column():
-            revenue_ttm = gr.Number(label="Revenue TTM (USD)", value=1_200_000)
-            revenue_growth_3y = gr.Slider(0, 5, step=0.01, value=2.3, label="Revenue Growth 3-Yr (x)")
-            burn_rate_monthly = gr.Number(label="Burn Rate (USD/month)", value=50_000)
-            cash_reserves = gr.Number(label="Cash Reserves (USD)", value=600_000)
-            profit_margin = gr.Slider(-1, 1, step=0.01, value=0.12, label="Profit Margin")
-            team_cohesion_score = gr.Slider(0, 1, step=0.01, value=0.85, label="Team Cohesion Score")
-            execution_velocity = gr.Slider(0, 1, step=0.01, value=0.78, label="Execution Velocity")
-            founder_commitment_score = gr.Slider(0, 1, step=0.01, value=0.90, label="Founder Commitment Score")
-            market_timing_score = gr.Slider(0, 1, step=0.01, value=0.70, label="Market Timing Score")
-            sector_heat_score = gr.Slider(0, 1, step=0.01, value=0.65, label="Sector Heat Score")
-            competitive_landscape_score = gr.Slider(0, 1, step=0.01, value=0.30, label="Competitive Landscape Score")
-            regulatory_environment_score = gr.Slider(0, 1, step=0.01, value=0.80, label="Regulatory Environment Score")
-            product_market_fit = gr.Slider(0, 1, step=0.01, value=0.75, label="Product-Market Fit Score")
-            transparency_score = gr.Slider(0, 1, step=0.01, value=0.85, label="Transparency Score")
-            compliance_score = gr.Slider(0, 1, step=0.01, value=0.82, label="Compliance Score")
-            board_independence = gr.Slider(0, 1, step=0.01, value=0.77, label="Board Independence Score")
-
-            btn = gr.Button("🔮 ประเมินมูลค่า", variant="primary")
-
+            inputs=[gr.Number(label="Revenue TTM (USD)",value=1_200_000),
+                    gr.Slider(0,5,step=0.01,value=2.3,label="Revenue Growth 3-Yr (x)"),
+                    gr.Number(label="Burn Rate (USD/m)",value=50_000),
+                    gr.Number(label="Cash Reserves (USD)",value=600_000),
+                    gr.Slider(-1,1,step=0.01,value=0.12,label="Profit Margin"),
+                    gr.Slider(0,1,step=0.01,value=0.85,label="Team Cohesion Score"),
+                    gr.Slider(0,1,step=0.01,value=0.78,label="Execution Velocity"),
+                    gr.Slider(0,1,step=0.01,value=0.90,label="Founder Commitment Score"),
+                    gr.Slider(0,1,step=0.01,value=0.70,label="Market Timing Score"),
+                    gr.Slider(0,1,step=0.01,value=0.65,label="Sector Heat Score"),
+                    gr.Slider(0,1,step=0.01,value=0.30,label="Competitive Landscape Score"),
+                    gr.Slider(0,1,step=0.01,value=0.80,label="Regulatory Environment Score"),
+                    gr.Slider(0,1,step=0.01,value=0.75,label="Product-Market Fit Score"),
+                    gr.Slider(0,1,step=0.01,value=0.85,label="Transparency Score"),
+                    gr.Slider(0,1,step=0.01,value=0.82,label="Compliance Score"),
+                    gr.Slider(0,1,step=0.01,value=0.77,label="Board Independence Score")]
+            btn=gr.Button("🔮 ประเมินมูลค่า",variant="primary")
         with gr.Column():
-            correlation = gr.Textbox(label="📊 Correlation", interactive=False)
-            sharpe = gr.Textbox(label="📈 Sharpe Ratio", interactive=False)
-            bvi_score = gr.Textbox(label="🎯 BVI Score", interactive=False)
-            risk_level = gr.Textbox(label="⚠️ Risk Level", interactive=False)
-            world_ranking = gr.Textbox(label="🌍 World Ranking", interactive=False)
-            projected_roi = gr.Textbox(label="💰 Projected ROI (x)", interactive=False)
-            pdf_file = gr.File(label="📄 ดาวน์โหลดรายงาน PDF", type="binary")
-
-    # event
-    btn.click(
-        evaluate,
-        inputs=list(locals().values())[:-7],  # ตัดตัวแปรหลัง btn ออก
-        outputs=[correlation, sharpe, bvi_score, risk_level, world_ranking, projected_roi, pdf_file],
-    )
-
-    # ปุ่มสุ่มตัวอย่าง
-    gr.Markdown("### 🎲 สุ่มตัวอย่างข้อมูลใหม่")
-    random_btn = gr.Button("สุ่มข้อมูล")
-    random_btn.click(
-        lambda: [gr.update(value=v) for v in dummy_data().iloc[0].tolist()[:-1]],  # ยกเว้น actual_roi
-        outputs=list(locals().values())[:-7],
-    )
-
-demo.queue(max_size=50).launch(server_name="0.0.0.0", server_port=7860, share=True)
+            outputs=[gr.Textbox(label="📊 Correlation",interactive=False),
+                     gr.Textbox(label="📈 Sharpe Ratio",interactive=False),
+                     gr.Textbox(label="🎯 BVI Score",interactive=False),
+                     gr.Textbox(label="⚠️ Risk Level",interactive=False),
+                     gr.Textbox(label="🌍 World Ranking",interactive=False),
+                     gr.Textbox(label="💰 Projected ROI (x)",interactive=False),
+                     gr.File(label="📄 ดาวน์โหลดรายงาน PDF",type="binary")]
+    btn.click(evaluate,inputs=inputs,outputs=outputs)
+    # ปุ่มสุ่ม
+    rng_btn=gr.Button("🎲 สุ่มข้อมูล")
+    def random_fill():
+        rng=np.random.default_rng(); n=1
+        df=pd.DataFrame({"revenue_ttm":rng.lognormal(14,1.5,n),
+                         "revenue_growth_3y":rng.beta(2,3,n)*5,
+                         "burn_rate_monthly":rng.exponential(50_000,n),
+                         "cash_reserves":rng.lognormal(15,1,n),
+                         "profit_margin":rng.normal(0.1,0.2,n),
+                         "team_cohesion_score":rng.beta(4,1.5,n),
+                         "execution_velocity":rng.beta(3,2,n),
+                         "founder_commitment_score":rng.beta(5,1,n),
+                         "market_timing_score":rng.beta(3,3,n),
+                         "sector_heat_score":rng.beta(2,2,n),
+                         "competitive_landscape_score":rng.beta(2,2,n),
+                         "regulatory_environment_score":rng.beta(3,1,n),
+                         "product_market_fit":rng.beta(2,2,n),
+                         "transparency_score":rng.beta(9,1,n),
+                         "compliance_score":rng.beta(9,1,n),
+                         "board_independence":rng.beta(8,2,n)})
+        return [v[0] for v in df.iloc[0].tolist()]
+    rng_btn.click(random_fill,outputs=inputs)
+
+# --------------- 7.  Launch ---------------
+demo.queue(max_size=50).launch(server_name="0.0.0.0",server_port=7860,share=True)