fix(ml): handle missing volume data and skip volume-dependent indicators

- Fill volume with 0 when column is absent from candle data
- Skip MFI/OBV/AD/ADOSC indicators when no real volume data available
- Fix pandas FutureWarning for inplace fillna in candle_features
- Remove temporary debug NaN logging

services/ml/app/preprocessing.py

@@ -60,6 +60,11 @@ def preprocess_candles(
     except Exception as e:
         raise ValueError(f"Candle data validation failed: {e}")
     
+    # Handle missing volume column - fill with 0 if absent
+    if 'volume' not in df.columns:
+        logger.warning("Volume column missing from candle data, filling with 0")
+        df['volume'] = 0.0
+    
     # Get feature engineering config
     fe_config = pipeline_config.stages.feature_engineering
     
@@ -69,9 +74,19 @@ def preprocess_candles(
     
     # Compute TA-Lib indicators
     if fe_config.talib_indicators:
-        logger.info(f"Computing {len(fe_config.talib_indicators)} TA-Lib indicators")
+        indicators = fe_config.talib_indicators
+        # Skip volume-dependent indicators when volume data is unavailable
+        volume_indicators = {'MFI', 'OBV', 'AD', 'ADOSC'}
+        has_real_volume = df['volume'].sum() > 0
+        if not has_real_volume:
+            skipped = [i.name for i in indicators if i.name.upper() in volume_indicators]
+            if skipped:
+                logger.warning(f"Skipping volume-dependent indicators (no volume data): {skipped}")
+            indicators = [i for i in indicators if i.name.upper() not in volume_indicators]
+        
+        logger.info(f"Computing {len(indicators)} TA-Lib indicators")
         try:
-            df = compute_talib_indicators(df, fe_config.talib_indicators)
+            df = compute_talib_indicators(df, indicators)
         except Exception as e:
             logger.error(f"Failed to compute TA-Lib indicators: {e}")
             raise ValueError(f"Indicator computation failed: {e}")

services/ml/features/candle_features.py

@@ -88,7 +88,7 @@ def compute_candle_features(df: pd.DataFrame) -> pd.DataFrame:
     # Gap (open - previous close)
     # For the first candle, gap is 0
     result_df['gap'] = result_df['open'] - result_df['close'].shift(1)
-    result_df['gap'].fillna(0.0, inplace=True)
+    result_df['gap'] = result_df['gap'].fillna(0.0)
     
     logger.info("Computed 8 candle features: body_size, body_direction, upper_wick, "
                 "lower_wick, wick_ratio, body_to_range, gap, range")