fez5587/commander-ai-lab
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(ml): enable ml_log=True in deepseek route engine calls

Browse Source
This commit is contained in:
KoalaTrapLord authored and GitHub committed 2026-03-21 12:41:01 -04:00
1 parent b1996a8deb
commit 963a5db638
1 file changed
+5 -5
Download Patch File Download Diff File Expand all files Collapse all files
routes/deepseek.py
+5 -5
View File
@@ -274,7 +274,7 @@ def _run_sim_thread(sim_id: str, decklist: list, num_games: int,
deck_a = build_deck(decklist)
deck_b = _generate_training_deck()
engine = GameEngine(max_turns=25, record_log=record_logs)
engine = GameEngine(max_turns=25, record_log=record_logs, ml_log=True, ml_log=True)
game_results, stats, elapsed = _run_game_loop(
engine, deck_a, deck_b, deck_name, 'Training Deck',
@@ -301,7 +301,7 @@ def _run_sim_thread_v2(sim_id: str, card_data: list[dict], num_games: int,
deck_a = _build_deck_from_card_data(card_data)
deck_b = _generate_training_deck()
engine = GameEngine(max_turns=25, record_log=record_logs)
engine = GameEngine(max_turns=25, record_log=record_logs, ml_log=True, ml_log=True)
game_results, stats, elapsed = _run_game_loop(
engine, deck_a, deck_b, deck_name, 'Training Deck',
@@ -337,7 +337,7 @@ def _run_sim_thread_deepseek(sim_id: str, card_data: list[dict],
validator = _get_validator_brain() # None if disabled
engine = DeepSeekGameEngine(
brain=brain, ai_player_index=1,
max_turns=25, record_log=record_logs,
max_turns=25, record_log=record_logs, ml_log=True,
validator=validator)
game_results, stats, elapsed = _run_game_loop(
@@ -391,9 +391,9 @@ def _run_sim_thread_n_player(sim_id: str, decks_data: list[list[dict]],
brain.check_connection()
engine = DeepSeekGameEngine(
brain=brain, ai_player_index=1,
max_turns=25, record_log=record_logs)
max_turns=25, record_log=record_logs, ml_log=True, ml_log=True)
else:
engine = GameEngine(max_turns=25, record_log=record_logs)
engine = GameEngine(max_turns=25, record_log=record_logs, ml_log=True, ml_log=True)
# Per-seat accumulators
seat_wins = [0] * n_players