Upload Quasar_axrvi_ranker.py

Quasar_axrvi_ranker.py

@@ -392,8 +392,8 @@ GAMMA           = 0.99
 LAMBDA_RANK     = 0.4
 LAMBDA_RISK     = 0.3
 REPLAY_CAPACITY = 10_000
-TRAIN_BATCH     = 8    # Lowered from 16 — fills after ~8 closed trades (~2-4 min)
-TRAIN_EVERY_N   = 5   # Lowered from 10 — checks buffer every ~25s
+TRAIN_BATCH     = 4    # FIX 2: Lowered from 8 — fills after ~4 closed trades (~1-2 min)
+TRAIN_EVERY_N   = 3   # FIX 2: Lowered from 5 — checks buffer every ~15s
 
 # Connection
 WS_RECONNECT_DELAY   = 5
@@ -3394,7 +3394,11 @@ class HybridTrainer:
                and "selected_idx" in ep and "pnl_per_asset" in ep
         ]
         if not valid:
-            logger.debug("[HybridTrainer] No valid episodes — skipping")
+            logger.warning(
+                f"[HybridTrainer] No valid episodes in batch of {len(episodes)} — "
+                f"skipping training step. Episodes may be missing required keys: "
+                f"'sequences', 'next_sequences', 'selected_idx', 'pnl_per_asset'."
+            )
             return {}
 
         self.model.train()
@@ -5280,8 +5284,17 @@ class QuasarAXRVIBridge:
         self.portfolio_risk_mgr.register_open(trade_id, quantity * price)
 
         # [S2] Capture s_t ∈ F_t at the moment of trade open
-        selected_idx = self._last_selected_idx or 0
-        self._open_pending_episode(trade_id, asset, selected_idx)
+        # FIX 3: Guard against None — do NOT silently default to index 0,
+        # which would attribute this episode to the wrong asset entirely.
+        if self._last_selected_idx is None:
+            logger.warning(
+                f"[_open_pending_episode] [{asset}] No selected_idx available — "
+                f"skipping episode capture for trade_id={trade_id}. "
+                f"Trade will still execute but will NOT contribute to training."
+            )
+        else:
+            selected_idx = self._last_selected_idx
+            self._open_pending_episode(trade_id, asset, selected_idx)
         # [S8] Initialise tick counter for optimal-stopping monitor
         self._trade_tick_counts[trade_id] = 0
 
@@ -7274,6 +7287,8 @@ if __name__ == "__main__":
                 model_path      = args.model,
                 hub_ws_url      = args.hub,
                 enable_logging  = not args.no_logs,
+                checkpoint_dir  = args.checkpoint_dir,   # FIX 1: was silently ignored
+                resume          = args.resume,            # FIX 1: was silently ignored
             ))
     except KeyboardInterrupt:
         print("\n👋 Shutting down…")