Method select

optillm/mcts.py:45–52

(self, node: MCTSNode)

            self.max_tokens = request_config.get('max_tokens', self.max_tokens)

    def select(self, node: MCTSNode) -> MCTSNode:
        logger.debug(f"Selecting node. Current node visits: {node.visits}, value: {node.value}")
        if not node.children:
            logger.debug("Node has no children. Returning current node.")
            return node
        selected_node = max(node.children, key=lambda c: c.value / (c.visits + 1e-8) + self.exploration_weight * np.sqrt(np.log(node.visits + 1) / (c.visits + 1e-8)))
        logger.debug(f"Selected child node. Visits: {selected_node.visits}, Value: {selected_node.value}")
        return selected_node

    def expand(self, node: MCTSNode) -> MCTSNode:
        logger.debug(f"Expanding node. Current state: {node.state}")