Spaces:

Alvin3y1
/

test

Sleeping

App Files Files Community

Alvin3y1 commited on 13 days ago

Commit

a4227fa

verified ·

1 Parent(s): aad8eb4

Update app.py

Browse files

Files changed (1) hide show

app.py +62 -43

app.py CHANGED Viewed

@@ -9,7 +9,7 @@ from aiohttp import web
 import websockets
 from sklearn.ensemble import RandomForestRegressor
-# Configuration
 SYMBOL_KRAKEN = "BTC/USD"
 PORT = 7860
 BROADCAST_RATE = 1.0
@@ -82,14 +82,12 @@ def calculate_indicators(candles):
 def train_model(df):
     logging.info("Training ML Model...")
-    # 1. Prepare Features (X)
     feature_cols = ['close', 'ema', 'bb_upper', 'bb_lower', 'rsi', 'macd', 'stoch_k', 'atr', 'obv', 'vwap']
-    # Create a clean copy for training data
     data = df.dropna().copy()
-    # 2. Create Targets (y) - OPTIMIZED to fix Fragmentation Warning
-    # Instead of adding columns in a loop, we create a dict and concat once
     future_shifts = {}
     targets = []
@@ -98,11 +96,9 @@ def train_model(df):
         future_shifts[col_name] = data['close'].shift(-i)
         targets.append(col_name)
-    # Concatenate all target columns at once
     target_df = pd.DataFrame(future_shifts, index=data.index)
     data = pd.concat([data, target_df], axis=1)
-    # Drop rows where we don't have future data (the last 100 rows)
     data = data.dropna()
     if len(data) < 100:
@@ -122,17 +118,13 @@ def train_model(df):
 def get_prediction(df, model):
     if model is None: return []
-    # Get the very last row of data (current market state)
     feature_cols = ['close', 'ema', 'bb_upper', 'bb_lower', 'rsi', 'macd', 'stoch_k', 'atr', 'obv', 'vwap']
     last_row = df.iloc[[-1]][feature_cols]
-    # Check for NaNs
     if last_row.isnull().values.any(): return []
-    # Predict
     prediction = model.predict(last_row.values)[0]
-    # Format for frontend
     current_time = int(df.iloc[-1]['time'])
     pred_data = []
     for i, price in enumerate(prediction):
@@ -152,16 +144,24 @@ def process_market_data():
     if df is None or len(df) < 50: return {"error": "Not enough data"}
     # 2. Train Model (Periodically)
-    # Train initially or every 15 minutes (900 seconds)
     if market_state['model'] is None or (time.time() - market_state['last_training_time'] > 900):
-        market_state['model'] = train_model(df)
-        market_state['last_training_time'] = time.time()
     # 3. Get Prediction
-    predictions = get_prediction(df, market_state['model'])
-    # 4. Prepare JSON
-    full_data = df.where(pd.notnull(df), None).to_dict('records')
     return {
         "data": full_data,
@@ -190,21 +190,21 @@ HTML_PAGE = f"""
 <body>
     <div class="header">
         <span style="color:#00e676">{SYMBOL_KRAKEN} + Random Forest (Next 100 Candles)</span>
-        <span id="clock" style="color:#888">Initializing...</span>
     </div>
     <div id="charts-container">
         <div id="main-chart" class="chart-row">
             <div class="legend">
                 <span class="l-item" style="color:#00ff9d">Price</span>
-                <span class="l-item" style="color:#bf5af2">AI Prediction</span>
                 <span class="l-item" style="color:#2962FF">EMA</span>
             </div>
         </div>
         <div id="osc-chart" class="chart-row">
             <div class="legend">
                 <span class="l-item" style="color:#9C27B0">RSI</span>
-                <span class="l-item" style="color:#00BCD4">MACD</span>
             </div>
         </div>
     </div>
@@ -222,16 +222,18 @@ HTML_PAGE = f"""
             crosshair: {{ mode: 1 }}
         }};
         const mainChart = LightweightCharts.createChart(mainEl, commonOpts);
         const candles = mainChart.addCandlestickSeries({{ upColor: '#00ff9d', downColor: '#ff3b3b', borderVisible: false }});
         const ema = mainChart.addLineSeries({{ color: '#2962FF', lineWidth: 1 }});
-        const predLine = mainChart.addLineSeries({{ color: '#bf5af2', lineWidth: 2, lineStyle: 2, title: 'AI Forecast' }});
         const oscChart = LightweightCharts.createChart(oscEl, commonOpts);
         const rsi = oscChart.addLineSeries({{ color: '#9C27B0', lineWidth: 1 }});
         const macdHist = oscChart.addHistogramSeries({{ priceScaleId: 'macd', color: '#2962FF' }});
         oscChart.priceScale('macd').applyOptions({{ scaleMargins: {{ top: 0.8, bottom: 0 }} }});
         new ResizeObserver(entries => {{
              for (let e of entries) {{
                  if(e.target === mainEl) mainChart.applyOptions({{ width: e.contentRect.width, height: e.contentRect.height }});
@@ -239,6 +241,7 @@ HTML_PAGE = f"""
              }}
         }}).observe(document.body);
         function syncCharts(source, targets) {{
             source.timeScale().subscribeVisibleLogicalRangeChange(range => {{
                 targets.forEach(t => t.timeScale().setVisibleLogicalRange(range));
@@ -247,33 +250,50 @@ HTML_PAGE = f"""
         syncCharts(mainChart, [oscChart]);
         syncCharts(oscChart, [mainChart]);
         function connect() {{
             const ws = new WebSocket((location.protocol === 'https:' ? 'wss' : 'ws') + '://' + location.host + '/ws');
             ws.onmessage = (e) => {{
-                const payload = JSON.parse(e.data);
-                if (!payload.data) return;
-                const d = payload.data;
-                const mapData = (key) => d.map(x => ({{ time: x.time, value: x[key] }})).filter(x => x.value !== null);
-                candles.setData(d.map(x => ({{ time: x.time, open: x.open, high: x.high, low: x.low, close: x.close }})));
-                ema.setData(mapData('ema'));
-                rsi.setData(mapData('rsi'));
-                if(payload.prediction && payload.prediction.length > 0) {{
-                    predLine.setData(payload.prediction);
                 }}
-                macdHist.setData(d.map(x => ({{
-                    time: x.time,
-                    value: x.macd_hist || 0,
-                    color: (x.macd_hist||0) >= 0 ? '#26a69a' : '#ef5350'
-                }})));
-                document.getElementById('clock').innerText = new Date().toISOString().split('T')[1].split('.')[0] + ' UTC';
             }};
-            ws.onclose = () => setTimeout(connect, 2000);
         }}
         connect();
     }});
 </script>
@@ -309,7 +329,6 @@ async def kraken_worker():
     except Exception as e:
         logging.error(f"Init Error: {e}")
-    # WebSocket Stream
     while True:
         try:
             async with websockets.connect("wss://ws.kraken.com/v2") as ws:

 import websockets
 from sklearn.ensemble import RandomForestRegressor
+# --- Configuration ---
 SYMBOL_KRAKEN = "BTC/USD"
 PORT = 7860
 BROADCAST_RATE = 1.0
 def train_model(df):
     logging.info("Training ML Model...")
     feature_cols = ['close', 'ema', 'bb_upper', 'bb_lower', 'rsi', 'macd', 'stoch_k', 'atr', 'obv', 'vwap']
+    # Clean data for training
     data = df.dropna().copy()
+    # Create Targets efficiently (fix for fragmentation warning)
     future_shifts = {}
     targets = []
         future_shifts[col_name] = data['close'].shift(-i)
         targets.append(col_name)
     target_df = pd.DataFrame(future_shifts, index=data.index)
     data = pd.concat([data, target_df], axis=1)
     data = data.dropna()
     if len(data) < 100:
 def get_prediction(df, model):
     if model is None: return []
     feature_cols = ['close', 'ema', 'bb_upper', 'bb_lower', 'rsi', 'macd', 'stoch_k', 'atr', 'obv', 'vwap']
     last_row = df.iloc[[-1]][feature_cols]
     if last_row.isnull().values.any(): return []
     prediction = model.predict(last_row.values)[0]
     current_time = int(df.iloc[-1]['time'])
     pred_data = []
     for i, price in enumerate(prediction):
     if df is None or len(df) < 50: return {"error": "Not enough data"}
     # 2. Train Model (Periodically)
     if market_state['model'] is None or (time.time() - market_state['last_training_time'] > 900):
+        try:
+            market_state['model'] = train_model(df)
+            market_state['last_training_time'] = time.time()
+        except Exception as e:
+            logging.error(f"Training failed: {e}")
     # 3. Get Prediction
+    predictions = []
+    try:
+        predictions = get_prediction(df, market_state['model'])
+    except Exception as e:
+        logging.error(f"Prediction failed: {e}")
+    # 4. Clean Data for JSON (Remove Infinity/NaN)
+    # This prevents the "blank graph" issue caused by invalid JSON
+    df_clean = df.replace([np.inf, -np.inf], np.nan)
+    full_data = df_clean.where(pd.notnull(df_clean), None).to_dict('records')
     return {
         "data": full_data,
 <body>
     <div class="header">
         <span style="color:#00e676">{SYMBOL_KRAKEN} + Random Forest (Next 100 Candles)</span>
+        <span id="clock" style="color:#888">Connecting...</span>
     </div>
     <div id="charts-container">
         <div id="main-chart" class="chart-row">
             <div class="legend">
                 <span class="l-item" style="color:#00ff9d">Price</span>
+                <span class="l-item" style="color:#bf5af2">AI Forecast</span>
                 <span class="l-item" style="color:#2962FF">EMA</span>
             </div>
         </div>
         <div id="osc-chart" class="chart-row">
             <div class="legend">
                 <span class="l-item" style="color:#9C27B0">RSI</span>
+                <span class="l-item" style="color:#26a69a">MACD</span>
             </div>
         </div>
     </div>
             crosshair: {{ mode: 1 }}
         }};
+        // 1. Initialize Charts
         const mainChart = LightweightCharts.createChart(mainEl, commonOpts);
         const candles = mainChart.addCandlestickSeries({{ upColor: '#00ff9d', downColor: '#ff3b3b', borderVisible: false }});
         const ema = mainChart.addLineSeries({{ color: '#2962FF', lineWidth: 1 }});
+        const predLine = mainChart.addLineSeries({{ color: '#bf5af2', lineWidth: 2, lineStyle: 2 }});
         const oscChart = LightweightCharts.createChart(oscEl, commonOpts);
         const rsi = oscChart.addLineSeries({{ color: '#9C27B0', lineWidth: 1 }});
         const macdHist = oscChart.addHistogramSeries({{ priceScaleId: 'macd', color: '#2962FF' }});
         oscChart.priceScale('macd').applyOptions({{ scaleMargins: {{ top: 0.8, bottom: 0 }} }});
+        // 2. Responsive Resize
         new ResizeObserver(entries => {{
              for (let e of entries) {{
                  if(e.target === mainEl) mainChart.applyOptions({{ width: e.contentRect.width, height: e.contentRect.height }});
              }}
         }}).observe(document.body);
+        // 3. Sync Time Scales
         function syncCharts(source, targets) {{
             source.timeScale().subscribeVisibleLogicalRangeChange(range => {{
                 targets.forEach(t => t.timeScale().setVisibleLogicalRange(range));
         syncCharts(mainChart, [oscChart]);
         syncCharts(oscChart, [mainChart]);
+        // 4. WebSocket Logic
         function connect() {{
             const ws = new WebSocket((location.protocol === 'https:' ? 'wss' : 'ws') + '://' + location.host + '/ws');
             ws.onmessage = (e) => {{
+                try {{
+                    const payload = JSON.parse(e.data);
+                    if (!payload.data) return;
+                    const d = payload.data;
+                    // Helper to map safely (avoids undefined/null crashes)
+                    const mapData = (key) => d
+                        .map(x => ({{ time: x.time, value: x[key] }}))
+                        .filter(x => x.value !== null && x.value !== undefined);
+                    // Set Data
+                    candles.setData(d.map(x => ({{ time: x.time, open: x.open, high: x.high, low: x.low, close: x.close }})));
+                    ema.setData(mapData('ema'));
+                    rsi.setData(mapData('rsi'));
+                    if(payload.prediction && payload.prediction.length > 0) {{
+                        predLine.setData(payload.prediction);
+                    }}
+                    macdHist.setData(d.map(x => ({{
+                        time: x.time,
+                        value: x.macd_hist || 0,
+                        color: (x.macd_hist||0) >= 0 ? '#26a69a' : '#ef5350'
+                    }})));
+                    // Update Clock
+                    document.getElementById('clock').innerText = new Date().toISOString().split('T')[1].split('.')[0] + ' UTC';
+                }} catch (err) {{
+                    console.error("Chart Render Error:", err);
                 }}
             }};
+            ws.onclose = () => {{
+                document.getElementById('clock').innerText = "Disconnected. Retrying...";
+                setTimeout(connect, 2000);
+            }};
         }}
         connect();
     }});
 </script>
     except Exception as e:
         logging.error(f"Init Error: {e}")
     while True:
         try:
             async with websockets.connect("wss://ws.kraken.com/v2") as ws: