better:优化回复逻辑，现在回复前会先思考，移除推理模型再回复中的使用，优化心流运行逻辑，优化思考时间计算逻辑，添加错误检测

src/plugins/memory_system/Hippocampus.py

@@ -697,6 +697,11 @@ class ParahippocampalGyrus:
         start_time = time.time()
         logger.info("[遗忘] 开始检查数据库...")
 
+        # 验证百分比参数
+        if not 0 <= percentage <= 1:
+            logger.warning(f"[遗忘] 无效的遗忘百分比: {percentage}, 使用默认值 0.005")
+            percentage = 0.005
+
         all_nodes = list(self.memory_graph.G.nodes())
         all_edges = list(self.memory_graph.G.edges())
 
@@ -704,11 +709,21 @@ class ParahippocampalGyrus:
             logger.info("[遗忘] 记忆图为空,无需进行遗忘操作")
             return
 
-        check_nodes_count = max(1, int(len(all_nodes) * percentage))
-        check_edges_count = max(1, int(len(all_edges) * percentage))
+        # 确保至少检查1个节点和边，且不超过总数
+        check_nodes_count = max(1, min(len(all_nodes), int(len(all_nodes) * percentage)))
+        check_edges_count = max(1, min(len(all_edges), int(len(all_edges) * percentage)))
 
-        nodes_to_check = random.sample(all_nodes, check_nodes_count)
-        edges_to_check = random.sample(all_edges, check_edges_count)
+        # 只有在有足够的节点和边时才进行采样
+        if len(all_nodes) >= check_nodes_count and len(all_edges) >= check_edges_count:
+            try:
+                nodes_to_check = random.sample(all_nodes, check_nodes_count)
+                edges_to_check = random.sample(all_edges, check_edges_count)
+            except ValueError as e:
+                logger.error(f"[遗忘] 采样错误: {str(e)}")
+                return
+        else:
+            logger.info("[遗忘] 没有足够的节点或边进行遗忘操作")
+            return
 
         # 使用列表存储变化信息
         edge_changes = {

src/plugins/memory_system/sample_distribution.py

@@ -58,8 +58,18 @@ class MemoryBuildScheduler:
             weight2 (float): 第二个分布的权重
             total_samples (int): 要生成的总时间点数量
         """
+        # 验证参数
+        if total_samples <= 0:
+            raise ValueError("total_samples 必须大于0")
+        if weight1 < 0 or weight2 < 0:
+            raise ValueError("权重必须为非负数")
+        if std_hours1 < 0 or std_hours2 < 0:
+            raise ValueError("标准差必须为非负数")
+
         # 归一化权重
         total_weight = weight1 + weight2
+        if total_weight == 0:
+            raise ValueError("权重总和不能为0")
         self.weight1 = weight1 / total_weight
         self.weight2 = weight2 / total_weight
 
@@ -73,12 +83,11 @@ class MemoryBuildScheduler:
     def generate_time_samples(self):
         """生成混合分布的时间采样点"""
         # 根据权重计算每个分布的样本数
-        samples1 = int(self.total_samples * self.weight1)
-        samples2 = self.total_samples - samples1
+        samples1 = max(1, int(self.total_samples * self.weight1))
+        samples2 = max(1, self.total_samples - samples1)  # 确保 samples2 至少为1
 
         # 生成两个正态分布的小时偏移
         hours_offset1 = np.random.normal(loc=self.n_hours1, scale=self.std_hours1, size=samples1)
-
         hours_offset2 = np.random.normal(loc=self.n_hours2, scale=self.std_hours2, size=samples2)
 
         # 合并两个分布的偏移