diff --git a/Decoder.py b/Decoder.py
index 3b1221a..d5e7805 100644
--- a/Decoder.py
+++ b/Decoder.py
@@ -425,7 +425,7 @@ class Decoder_main(threading.Thread):
                 y_pred = torch.max(Cls, 1)[1]
                 self.plotLabel.append(int(y_pred.item()))
                 algo_log(f"MI运动意图识别: {y_pred}")
-            self.zmqServer.broadcast_message('paradigm', int(y_pred.item()))
+            self.zmqServer.broadcast_message('result', int(y_pred.item()))
             end = time.time()
             algo_log(f'MI发送给界面完成,耗时{end - start:.3f}s。')
         else:  # 休息状态
diff --git a/MI/Algorithm/conformer_2class.py b/MI/Algorithm/conformer_2class.py
index 69a77f3..8efb0c8 100644
--- a/MI/Algorithm/conformer_2class.py
+++ b/MI/Algorithm/conformer_2class.py
@@ -318,11 +318,7 @@ class ExP():
                 train_pred = torch.max(outputs, 1)[1]
                 train_acc = float((train_pred == label).cpu().numpy().astype(int).sum()) / float(label.size(0))
 
-                algo_log('Epoch:', e,
-                      '  Train loss: %.6f' % loss.detach().cpu().numpy(),
-                      '  Test loss: %.6f' % loss_test.detach().cpu().numpy(),
-                      '  Train accuracy %.6f' % train_acc,
-                      '  Test accuracy is %.6f' % acc, level="debug")
+                algo_log(f"Epoch = {e}, Train loss = {loss.detach().cpu().numpy():.6f}, Test loss = {loss_test.detach().cpu().numpy():.6f}, Train accuracy = {train_acc:.6f}, Test accuracy = {acc:.6f}", level="debug")
 
                 self.log_write.write(str(e) + "    " + str(acc) + "\n")
                 num = num + 1
@@ -335,8 +331,8 @@ class ExP():
 
         torch.save(self.model, model_path)
         averAcc = averAcc / num
-        algo_log('The average accuracy is:', averAcc, level="debug")
-        algo_log('The best accuracy is:', bestAcc, level="debug")
+        algo_log(f"The average accuracy is: {averAcc}", level="debug")
+        algo_log(f"The best accuracy is: {bestAcc}", level="debug")
         self.log_write.write('The average accuracy is: ' + str(averAcc) + "\n")
         self.log_write.write('The best accuracy is: ' + str(bestAcc) + "\n")
 
@@ -366,12 +362,13 @@ def onlineTrain(data_queue,result_queue):
         data = data_queue.get(timeout=30)
         all_data, all_label,model_path,n_chan = data['data'], data['label'],data['modelPath'],data['n_chan']
         exp = ExP(n_chan)
-        algo_log('训练参数： ',np.shape(all_data),np.shape(all_label),model_path, level="debug")
+        algo_log(f"训练参数： {np.shape(all_data)}, {np.shape(all_label)}, {model_path}", level="debug")
         bestAcc, averAcc, Y_true, Y_pred = exp.train(all_data,all_label,model_path)
-        algo_log('THE BEST ACCURACY IS ' + str(bestAcc), level="debug")
+        algo_log(f"THE BEST ACCURACY IS {str(bestAcc)}", level="debug")
 
         endtime = datetime.datetime.now()
-        algo_log('train duration: ',str(endtime - starttime), level="debug")
+        algo_log(f"train duration: {endtime - starttime}", level="debug")
+        
 
         # 将模型或参数传回
         result_queue.put({
@@ -387,7 +384,7 @@ def offlineTrain(all_data,all_label,modelPath):
 
     # seed_n = np.random.randint(2025)
     seed_n = 1877
-    algo_log('seed is ' + str(seed_n), level="debug")
+    algo_log(f"seed is {seed_n}", level="debug")
     random.seed(seed_n)
     np.random.seed(seed_n)
     torch.manual_seed(seed_n)
@@ -400,7 +397,7 @@ def offlineTrain(all_data,all_label,modelPath):
     algo_log('THE BEST ACCURACY IS ' + str(bestAcc), level="debug")
 
     endtime = datetime.datetime.now()
-    algo_log('train duration: ',str(endtime - starttime), level="debug")
+    algo_log(f"train duration: {endtime - starttime}", level="debug")
 
 
 
diff --git a/SSVEP/dwfbcca.py b/SSVEP/dwfbcca.py
index b8474ad..4a2cfaa 100644
--- a/SSVEP/dwfbcca.py
+++ b/SSVEP/dwfbcca.py
@@ -18,11 +18,11 @@ from logs.log import algo_log
 class FbccaDw:
     def __init__(self, fs, num_target, num_chans, num_filter, num_harms, stimTime, parameter, width, winNum,method):
         algo_log('******************************************', level="debug")
-        algo_log('parameter list', level="debug")
-        algo_log('target:', num_target, level="debug")
-        algo_log('number of filter bank:', num_filter, level="debug")
-        algo_log('parameter:', parameter, level="debug")
-        algo_log('width:', width, level="debug")
+        algo_log('parameter list',level="debug")
+        algo_log(f"target: {num_target}", level="debug")
+        algo_log(f"number of filter bank: {num_filter}", level="debug")
+        algo_log(f"parameter: {parameter}", level="debug")
+        algo_log(f"width: {width}", level="debug")
         self.phase = 0
         self.bandWidth = width
         self.winNum = winNum
diff --git a/Zmq/zmqServer.py b/Zmq/zmqServer.py
index ac133c2..85db52e 100644
--- a/Zmq/zmqServer.py
+++ b/Zmq/zmqServer.py
@@ -152,7 +152,10 @@ class zmqServer(threading.Thread):
                 msg = {'method': method, 'params': params}
                 msg_bytes = json.dumps(msg).encode('utf-8')
 
-                algo_log(f"发送命令结果: {msg}", level="DEBUG")
+                if msg['method'] == 'beta_psd':
+                    algo_log(f"发送命令结果: {msg}", level="DEBUG", record_once=True)
+                else:
+                    algo_log(f"发送命令结果: {msg}", level="DEBUG")
                 
                 # 广播到所有命令客户端
                 for client_id in list(self.cmd_clients):