diff --git a/README.md b/README.md
index 437fb7e3850eda4d44c1f84f7ed5a633cb43833c..7ce63eb821b800a4ff41453f534847e0f8b9ab66 100644
--- a/README.md
+++ b/README.md
@@ -339,7 +339,10 @@ python StreamLearn/tests/test_ODS.py --data PATH_TO_DATA --checkpoint PATH_TO_CH
 - StreamLearn/Algorithm/Algorithm_SAFC/SAFC_Stream_Funcs.py
 
 
-首先，获取算法所需数据集：
+首先，根据代码配置文件StreamLearn/Algorithm/Algorithm_SAFC/request_import.py安装库完成环境配置
+
+
+然后，获取算法所需数据集：
 
 地址：
 通过百度网盘分享的文件：SAFC_datasets_CIFAR10.zip
@@ -347,84 +350,300 @@ python StreamLearn/tests/test_ODS.py --data PATH_TO_DATA --checkpoint PATH_TO_CH
 提取码：nudt
 
 说明：
-本实验需要用到的数据集是CIFAR10，本例将data_batch_1.mat至data_batch_4.mat中的数据作为第一阶段数据；data_batch_2.mat中的数据作为第二阶段的数据；test_batch.mat中的数据是测试数据
-
+本实验需要用到的数据集是CIFAR10，将部分batch中的数据作为第一阶段数据；部分数据作为第二阶段的数据；test_batch.mat中的数据是测试数据
 
-然后，根据代码配置文件StreamLearn/Algorithm/Algorithm_SAFC/request_import.py安装库完成环境配置
+流式数据读取，传入数据地址：
+```python
+def __init__(self,args_address):
+    # 存储地址
+    self.save_dir=args_address.save_dir
+        
+    # 读取地址
+    self.PathSet=args_address.PathSet
+    self.new_path=args_address.new_path
+    self.test_path=args_address.test_path
+```
 
 
 其次，调用SAFC算法进行训练：
-SAFC训练基于第一阶段复用的SVM模型，因此先训练SVM，并保存训练好的SVM模型：
-```python
-print('####eval####')
-print("begin svm1 training!")
-print('####eval####')
-svm1 = SVC(probability=True,kernel="linear",decision_function_shape='ovo')
-svm1.fit(data_s1.tolist(), label_s1_vec.tolist())
-dump(svm1, save_dir+'/svm1.model')
-print('####eval####')
-print("end svm1 training!")
-print('####eval####')
-
-# 复用SVM，训练SAFC_D，并保存训练后的模型：
-print('####eval####')
-print("begin SAFC_D training!")
-print('####eval####')
-w_ours1 = SAFC_D(np.mat(svm1._get_coef()), np.transpose(data_s2), label_s2, alpha_best1, beta_best1, eta)
-dump(w_ours1, save_dir+'/SAFC_D.model')
-
-# 复用SVM，训练SAFC_ID，并保存训练后的模型：
-print('####eval####')
-print("begin SAFC_ID training!")
-print('####eval####')
-w_ours2 = SAFC_ID(np.mat(svm1._get_coef()),np.transpose(data_s2),label_s2,alpha_best2,beta_best2,eta)
-dump(w_ours2, save_dir+'/SAFC_ID.model')
+```python
+def stream_fit(self):
+    # 第一阶段数据读取与训练
+    SAFC_Stage1(self.PathSet,self.save_dir)
+    # 第一阶段模型提取
+    print('####eval####')
+    print("Read stage1 model!")
+    print('####eval####')
+    stage1model=load(self.save_dir+'/svm1.model')
+    print('####eval####')
+    print("Have read stage1 model!")
+    print('####eval####')
+    
+    # 第二阶段数据读取与训练
+    SAFC_Stage2(self.new_path,stage1model,self.save_dir)
+    # 第二阶段模型提取
+    # 变体一
+    print('####eval####')
+    print("Read stage2-SAFC_D model!")
+    print('####eval####')
+    stage2model1=load(self.save_dir+'/SAFC_D.model')
+    print('####eval####')
+    print("Have read stage2-SAFC_D model!")
+    print('####eval####')
+    # 变体二
+    print('####eval####')
+    print("Read stage2-SAFC_D model!")
+    print('####eval####')
+    stage2model2=load(self.save_dir+'/SAFC_ID.model')
+    print('####eval####')
+    print("Have read stage2-SAFC_D model!")
+    print('####eval####')
+```
+
+具体而言，SAFC训练基于第一阶段复用的SVM模型，因此先训练SVM，并保存训练好的SVM模型：
+```python
+def SAFC_Stage1(PathSet,save_dir):
+    # 第一阶段数据读取与模型训练
+    print('####eval####')
+    print("开始读取第一阶段数据！")
+    print('####eval####')
+    
+    X_past,Y_past=readbatchtoPython(PathSet)
+    X_past,Y_past=datareconsrtuct(X_past,Y_past)
+    Y_past=Y_past+1
+
+    print('####eval####')
+    print("第一阶段数据读取完成！")
+    print('####eval####')
+
+    # Given and transform
+    data_s1=X_past
+    label_s1_vec=Y_past
+    label_s1 = ind2vec (Y_past.T, len(np.unique(Y_past)))
+
+    # 归一化
+    print('####eval####')
+    print("第一阶段数据归一化！")
+    print('####eval####')
+    scaler = StandardScaler()
+    data_s1 = scaler.fit_transform(data_s1)
+    print('####eval####')
+    print("第一阶段数据归一化完成！")
+    print('####eval####')
+
+    # 第一阶段数据特征输出
+    print('####eval####')
+    print("第一阶段数据：数据量：{}，特征维度：{}，类别数：{}".format(np.shape(data_s1)[0], np.shape(data_s1)[1], len(np.unique(Y_past))))
+    print('####eval####')
+
+    # 第一阶段模型训练与存储，用SVM训练第一阶段模型
+    print('####eval####')
+    print("begin svm1 training!")
+    print('####eval####')
+    svm1 = LinearSVC()
+    svm1.fit(data_s1.tolist(), label_s1_vec.tolist())
+    dump(svm1, save_dir+'/svm1.model')
+    print('####eval####')
+    print("end svm1 training!")
+    print('####eval####')
+    
+    return
+```
+
+复用SVM，在第二阶段训练SAFC_D和SAFC_ID两个变体，并保存训练后的模型：
+```python
+def SAFC_Stage2(new_path,svm1,save_dir):
+    # Paras
+    eta = 0.1
+    alpha_set = [10 ** -2, 10 ** -1, 10 ** 0]
+    alpha = 0.001
+    beta = 0.01  
+    
+    # 第二阶段数据读取与模型训练
+    print('####eval####')
+    print("开始读取第二阶段数据！")
+    print('####eval####')
+
+    X_new,Y_new=readtopython(new_path)
+    X_new,Y_new=datareconsrtuct(X_new,Y_new)
+    Y_new=Y_new+1
+
+    print('####eval####')
+    print("第二阶段数据读取完成！")
+    print('####eval####')
+
+    # Given and transform
+    data_s2=X_new
+    label_s2_vec=Y_new
+    label_s2 = ind2vec (Y_new.T, len(np.unique(Y_new)))
+
+    # 归一化
+    print('####eval####')
+    print("第二阶段数据归一化！")
+    print('####eval####')
+    scaler = StandardScaler()
+    data_s2 = scaler.fit_transform(data_s2)
+    print('####eval####')
+    print("第二阶段数据归一化完成！")
+    print('####eval####')
+
+    # 第二阶段数据特征输出
+    print('####eval####')
+    print("第二阶段涉及特征新增与类别新增！")
+    print('####eval####')
+    print('####eval####')
+    print("第二阶段数据：数据量：{}，特征维度：{}，类别数：{}".format(np.shape(data_s2)[0], np.shape(data_s2)[1], len(np.unique(Y_new))))
+    print('####eval####')
+
+    # 第二阶段模型训练与存储
+    # 变体一
+    alpha_best1, beta_best1 = validfunc(np.transpose(data_s2), label_s2, alpha_set, eta, "SAFC_D",np.mat(svm1.coef_))
+    startours1 = time.time()
+    print('####eval####')
+    print("begin SAFC_D training!")
+    print('####eval####')
+    w_ours1 = SAFC_D(np.mat(svm1.coef_), np.transpose(data_s2), label_s2, alpha_best1, beta_best1, eta)
+    dump(w_ours1, save_dir+'/SAFC_D.model')
+    timeours1 = time.time() - startours1
+    print('####eval####')
+    print(timeours1)
+    print('####eval####')
+    print('####eval####')
+    print("end SAFC_D training!")
+    print('####eval####')
+
+    # 变体二
+    alpha_best2,beta_best2=validfunc(np.transpose(data_s2),label_s2,alpha_set,eta,"SAFC_ID",np.mat(svm1.coef_))
+    startours2=time.time()
+    print('####eval####')
+    print("begin SAFC_ID training!")
+    print('####eval####')
+    w_ours2 = SAFC_ID(np.mat(svm1.coef_),np.transpose(data_s2),label_s2,alpha_best2,beta_best2,eta)
+    dump(w_ours2, save_dir+'/SAFC_ID.model')
+    timeours2=time.time()-startours2
+    print('####eval####')
+    print(timeours2)
+    print('####eval####')
+    print('####eval####')
+    print("end SAFC_ID training!")
+    print('####eval####')
+    
+    return
 
 # 其中，alpha_best1, beta_best1, alpha_best2, beta_best2, eta是超参数
 ```
 
 
-最后，分别对两个模型进行性能测试：
+最后，读取测试数据并完成测试与性能评估：
 ```python
-# SAFC_D：
-print('####eval####')
-print("begin SAFC_D testing!")
-print('####eval####')
-pred1, acc_ours1, auc_ours1, f1wei_ours1, f1macro_ours1, f1micro_ours1 = Predict(w_ours1,np.transpose(test_data),test_label)
-print('####eval####')
-print("begin SAFC_D evaluation!")
-print('####eval####')
-Acc_ours1.append(acc_ours1)
-AUC_ours1.append(auc_ours1)
-F1_weight_ours1.append(f1wei_ours1)
-F1_macro_ours1.append(f1macro_ours1)
-F1_micro_ours1.append(f1micro_ours1)
-
-# SAFC_ID：
-print('####eval####')
-print("begin SAFC_ID testing!")
-print('####eval####')
-pred2,acc_ours2,auc_ours2,f1wei_ours2,f1macro_ours2,f1micro_ours2 = Predict(w_ours2,np.transpose(test_data),test_label)
-print('####eval####')
-print("begin SAFC_ID evaluation!")
-print('####eval####')
-Acc_ours2.append(acc_ours2)
-AUC_ours2.append(auc_ours2)
-F1_weight_ours2.append(f1wei_ours2)
-F1_macro_ours2.append(f1macro_ours2)
-F1_micro_ours2.append(f1micro_ours2)
+def stream_evaluate(self):
+    # 测试数据读取与评估
+    SAFC_test(self.test_path,stage2model1,stage2model2)
 ```
 
-
-在test文件中，按照不同的数据流场景，提供两个main文件
-- StreamLearn/tests/tests_SAFC/CIFAR10_SAFC_DandSAFC_ID_allFunc_main.py
-对应一次性读取两阶段数据并完成测试
-- StreamLearn/tests/tests_SAFC/CIFAR10_SAFC_DandSAFC_ID_useFunc_main.py
-对应分批次读取两阶段数据，存储模型，读取测试数据，调用模型并完成测试
+具体而言，
+```python
+def SAFC_test(test_path,w_ours1,w_ours2):
+    # Metrics
+    Acc_ours1, Acc_ours2= [], []
+    AUC_ours1, AUC_ours2 = [], []
+    F1_weight_ours1, F1_weight_ours2 = [], []
+    F1_macro_ours1, F1_macro_ours2 = [], []
+    F1_micro_ours1, F1_micro_ours2 = [], []
+    
+    # 测试数据读取与评估
+    print('####eval####')
+    print("开始读取测试数据！")
+    print('####eval####')
+
+    X_test,Y_test=readtopython(test_path)
+    X_test,Y_test=datareconsrtuct(X_test,Y_test)
+    Y_test=Y_test+1
+
+    print('####eval####')
+    print("测试数据读取完成！")
+    print('####eval####')
+
+    # Given and transform
+    test_data=X_test
+    test_label_vec=Y_test
+    test_label = ind2vec (Y_test.T, len(np.unique(Y_test)))
+
+    # 归一化
+    print('####eval####')
+    print("测试数据归一化！")
+    print('####eval####')
+    scaler = StandardScaler()
+    test_data = scaler.fit_transform(test_data)
+    print('####eval####')
+    print("测试数据归一化完成！")
+    print('####eval####')
+
+    # 测试数据特征输出
+    print('####eval####')
+    print("测试数据：数据量：{}，特征维度：{}，类别数：{}".format(np.shape(test_data)[0], np.shape(test_data)[1], len(np.unique(Y_test))))
+    print('####eval####')
+
+    # Evaluation
+    # 变体一
+    print('####eval####')
+    print("begin SAFC_D testing!")
+    print('####eval####')
+    pred1, acc_ours1, auc_ours1, f1wei_ours1, f1macro_ours1, f1micro_ours1 = Predict(w_ours1,np.transpose(test_data),test_label)
+    print('####eval####')
+    print("begin SAFC_D evaluation!")
+    print('####eval####')
+    Acc_ours1.append(acc_ours1)
+    AUC_ours1.append(auc_ours1)
+    F1_weight_ours1.append(f1wei_ours1)
+    F1_macro_ours1.append(f1macro_ours1)
+    F1_micro_ours1.append(f1micro_ours1)
+
+    # 变体二
+    print('####eval####')
+    print("begin SAFC_ID testing!")
+    print('####eval####')
+    pred2,acc_ours2,auc_ours2,f1wei_ours2,f1macro_ours2,f1micro_ours2 = Predict(w_ours2,np.transpose(test_data),test_label)
+    print('####eval####')
+    print("begin SAFC_ID evaluation!")
+    print('####eval####')
+    Acc_ours2.append(acc_ours2)
+    AUC_ours2.append(auc_ours2)
+    F1_weight_ours2.append(f1wei_ours2)
+    F1_macro_ours2.append(f1macro_ours2)
+    F1_micro_ours2.append(f1micro_ours2)
+
+    # 结果输出
+    print('####eval####')
+    print("预测与评估完成，输出评估结果！")
+    print('####eval####')
+
+    meanAcc_ours1 = np.mean(Acc_ours1)
+    meanAcc_ours2 = np.mean(Acc_ours2)
+    meanAuc_ours1 = np.mean(AUC_ours1)
+    meanAuc_ours2 = np.mean(AUC_ours2)
+    meanF1_macro_ours1 = np.mean(F1_macro_ours1)
+    meanF1_macro_ours2 = np.mean(F1_macro_ours2)
+    meanF1_weight_ours1 = np.mean(F1_weight_ours1)
+    meanF1_weight_ours2 = np.mean(F1_weight_ours2)
+    meanF1_micro_ours1 = np.mean(F1_micro_ours1)
+    meanF1_micro_ours2 = np.mean(F1_micro_ours2)
+
+    print('####eval####')
+    print(
+        'meanAcc_ours1{:.4f},meanAcc_ours2{:.4f},meanAuc_ours1{:.4f},meanAuc_ours2{:.4f},meanF1_macro_ours1{:.4f},meanF1_macro_ours2{:.4f},meanF1_weight_ours1{:.4f},meanF1_weight_ours2{:.4f},meanF1_micro_ours1{:.4f},meanF1_micro_ours2{:.4f}'
+        .format(meanAcc_ours1,meanAcc_ours2, meanAuc_ours1,meanAuc_ours2,meanF1_macro_ours1,meanF1_macro_ours2,meanF1_weight_ours1,meanF1_weight_ours2,meanF1_micro_ours1,meanF1_micro_ours2))
+    print('####eval####')
+    print('Finished!')
+    
+    return
+```
 
 
 ModelsIN用于存放训练获得的分类器
 
+测试主文件为test_SAFC.py
+
 
 ## 课题四
 
diff --git a/StreamLearn/Algorithm/Algorithm_SAFC/Class_CIFAR10_SAFC_DandSAFC_ID.py b/StreamLearn/Algorithm/Algorithm_SAFC/Class_CIFAR10_SAFC_DandSAFC_ID.py
new file mode 100644
index 0000000000000000000000000000000000000000..0bf55c0025583557cadaf24c099cfaf5ce1eea04
--- /dev/null
+++ b/StreamLearn/Algorithm/Algorithm_SAFC/Class_CIFAR10_SAFC_DandSAFC_ID.py
@@ -0,0 +1,83 @@
+# -*- coding: utf-8 -*-
+"""
+Created on Sat Sep 14 2024
+
+@author: zhangxinyue
+"""
+
+import time
+
+import numpy as np
+import math
+
+from sklearn.preprocessing import StandardScaler
+from sklearn import model_selection
+from sklearn.metrics import *
+
+from sklearn import svm
+from sklearn.svm import SVC
+from sklearn.svm import LinearSVC
+
+import scipy.io as io
+from scipy.io import loadmat
+import os
+from joblib import dump,load
+
+from types import SimpleNamespace
+
+#Ours
+from StreamLearn.Algorithm.Algorithm_SAFC.compared_method import *
+from StreamLearn.Algorithm.Algorithm_SAFC.predict import *
+from StreamLearn.Algorithm.Algorithm_SAFC.SAFC import *
+from StreamLearn.Algorithm.Algorithm_SAFC.MetricsPred import *
+from StreamLearn.Algorithm.Algorithm_SAFC.validfunc_both import *
+from StreamLearn.Algorithm.Algorithm_SAFC.Read_to_Python import *
+from StreamLearn.Algorithm.Algorithm_SAFC.SAFC_Stream_Funcs import *
+
+class SAFC_achieve():
+    def __init__(self,args_address):
+        # %% 第一阶段、第二阶段、测试main函数
+        # 存储地址
+        self.save_dir=args_address.save_dir
+        
+        # 读取地址
+        self.PathSet=args_address.PathSet
+        self.new_path=args_address.new_path
+        self.test_path=args_address.test_path
+
+    def stream_fit(self):
+        # 第一阶段数据读取与训练
+        SAFC_Stage1(self.PathSet,self.save_dir)
+        # 第一阶段模型提取
+        print('####eval####')
+        print("Read stage1 model!")
+        print('####eval####')
+        stage1model=load(self.save_dir+'/svm1.model')
+        print('####eval####')
+        print("Have read stage1 model!")
+        print('####eval####')
+        
+        # 第二阶段数据读取与训练
+        SAFC_Stage2(self.new_path,stage1model,self.save_dir)
+        # 第二阶段模型提取
+        # 变体一
+        print('####eval####')
+        print("Read stage2-SAFC_D model!")
+        print('####eval####')
+        stage2model1=load(self.save_dir+'/SAFC_D.model')
+        print('####eval####')
+        print("Have read stage2-SAFC_D model!")
+        print('####eval####')
+        # 变体二
+        print('####eval####')
+        print("Read stage2-SAFC_D model!")
+        print('####eval####')
+        stage2model2=load(self.save_dir+'/SAFC_ID.model')
+        print('####eval####')
+        print("Have read stage2-SAFC_D model!")
+        print('####eval####')
+    
+    def stream_evaluate(self):
+        # 测试数据读取与评估
+        SAFC_test(self.test_path,stage2model1,stage2model2)
+
diff --git a/StreamLearn/tests/tests_SAFC/ModelsIN/.keep b/StreamLearn/Algorithm/Algorithm_SAFC/ModelsIN/.keep
similarity index 100%
rename from StreamLearn/tests/tests_SAFC/ModelsIN/.keep
rename to StreamLearn/Algorithm/Algorithm_SAFC/ModelsIN/.keep
diff --git a/StreamLearn/Algorithm/Algorithm_SAFC/Read_to_Python.py b/StreamLearn/Algorithm/Algorithm_SAFC/Read_to_Python.py
index c8486ba84ad3b782e7aa885fcafb5c360f29cfae..49ba73d973e511232a7f64d4cef2a75ee5378dd4 100644
--- a/StreamLearn/Algorithm/Algorithm_SAFC/Read_to_Python.py
+++ b/StreamLearn/Algorithm/Algorithm_SAFC/Read_to_Python.py
@@ -21,7 +21,8 @@ def readtopython(path):
     return X_part,Y_part
     
 # Batch数据接口
-def readbatchtoPython(Path):
+# Stage 1
+def readbatchtoPython_Past(Path):
     for path_index in range(len(Path)):
         path=Path[path_index]
         X_part,Y_part=readtopython(path)
@@ -38,4 +39,19 @@ def readbatchtoPython(Path):
     X_past=X_past_original[Position_past,0:2048]
     return X_past,Y_past
 
+# Stage 2
+def readbatchtoPython_New(Path):
+    for path_index in range(len(Path)):
+        path=Path[path_index]
+        X_part,Y_part=readtopython(path)
+        if path_index==0:
+            X_new_original=X_part
+            Y_new_original=Y_part
+        if path_index!=0:
+            X_new_original=np.vstack((X_new_original,X_part))
+            Y_new_original=np.vstack((Y_new_original,Y_part))
+    X_new=X_new_original
+    Y_new=Y_new_original
+    return X_new,Y_new
+
 
diff --git a/StreamLearn/Algorithm/Algorithm_SAFC/SAFC_Stream_Funcs.py b/StreamLearn/Algorithm/Algorithm_SAFC/SAFC_Stream_Funcs.py
index 343167d277f31ab48bdc74f909b7202c146a2858..4f7f4b4be787f58b1989b2aa37b6d371de4c8501 100644
--- a/StreamLearn/Algorithm/Algorithm_SAFC/SAFC_Stream_Funcs.py
+++ b/StreamLearn/Algorithm/Algorithm_SAFC/SAFC_Stream_Funcs.py
@@ -14,14 +14,6 @@ from sklearn.preprocessing import StandardScaler
 from sklearn import model_selection
 from sklearn.metrics import *
 
-from compared_method import *
-from predict import *
-from SAFC import *
-from MetricsPred import *
-from validfunc_both import *
-from Read_to_Python import *
-from SAFC_Stream_Funcs import *
-
 from sklearn import svm
 from sklearn.svm import SVC
 from sklearn.svm import LinearSVC
@@ -31,6 +23,16 @@ from scipy.io import loadmat
 import os
 from joblib import dump,load
 
+from types import SimpleNamespace
+
+#Ours
+from StreamLearn.Algorithm.Algorithm_SAFC.compared_method import *
+from StreamLearn.Algorithm.Algorithm_SAFC.predict import *
+from StreamLearn.Algorithm.Algorithm_SAFC.SAFC import *
+from StreamLearn.Algorithm.Algorithm_SAFC.MetricsPred import *
+from StreamLearn.Algorithm.Algorithm_SAFC.validfunc_both import *
+from StreamLearn.Algorithm.Algorithm_SAFC.Read_to_Python import *
+
 # %% Adress and Pre-define
 # 定义一个稀疏矩阵的方法
 def ind2vec(indices, num_classes):
@@ -56,7 +58,7 @@ def SAFC_Stage1(PathSet,save_dir):
     print("开始读取第一阶段数据！")
     print('####eval####')
     
-    X_past,Y_past=readbatchtoPython(PathSet)
+    X_past,Y_past=readbatchtoPython_Past(PathSet)
     X_past,Y_past=datareconsrtuct(X_past,Y_past)
     Y_past=Y_past+1
 
@@ -110,7 +112,7 @@ def SAFC_Stage2(new_path,svm1,save_dir):
     print("开始读取第二阶段数据！")
     print('####eval####')
 
-    X_new,Y_new=readtopython(new_path)
+    X_new,Y_new=readbatchtoPython_New(new_path)
     X_new,Y_new=datareconsrtuct(X_new,Y_new)
     Y_new=Y_new+1
 
diff --git a/StreamLearn/Algorithm/Algorithm_SAFC/request_import.py b/StreamLearn/Algorithm/Algorithm_SAFC/request_import.py
index 071fdbad0102cf41a4e0b0e38a72938ea8786474..5ac7cc99dddb617ce6ac6d3c75d9095e28f44b59 100644
--- a/StreamLearn/Algorithm/Algorithm_SAFC/request_import.py
+++ b/StreamLearn/Algorithm/Algorithm_SAFC/request_import.py
@@ -15,14 +15,6 @@ from sklearn.preprocessing import StandardScaler
 from sklearn import model_selection
 from sklearn.metrics import *
 
-from compared_method import *
-from predict import *
-from SAFC import *
-from MetricsPred import *
-from validfunc_both import *
-from Read_to_Python import *
-from SAFC_Stream_Funcs import *
-
 from sklearn import svm
 from sklearn.svm import SVC
 from sklearn.svm import LinearSVC
@@ -30,4 +22,16 @@ from sklearn.svm import LinearSVC
 import scipy.io as io
 from scipy.io import loadmat
 import os
-from joblib import dump,load
\ No newline at end of file
+from joblib import dump,load
+
+from types import SimpleNamespace
+
+#Ours
+from StreamLearn.Algorithm.Algorithm_SAFC.compared_method import *
+from StreamLearn.Algorithm.Algorithm_SAFC.predict import *
+from StreamLearn.Algorithm.Algorithm_SAFC.SAFC import *
+from StreamLearn.Algorithm.Algorithm_SAFC.MetricsPred import *
+from StreamLearn.Algorithm.Algorithm_SAFC.validfunc_both import *
+from StreamLearn.Algorithm.Algorithm_SAFC.Read_to_Python import *
+from StreamLearn.Algorithm.Algorithm_SAFC.SAFC_Stream_Funcs import *
+from StreamLearn.Algorithm.Algorithm_SAFC.Class_CIFAR10_SAFC_DandSAFC_ID import *
\ No newline at end of file
diff --git a/StreamLearn/tests/test_SAFC.py b/StreamLearn/tests/test_SAFC.py
new file mode 100644
index 0000000000000000000000000000000000000000..916ff878961d01b43a9e78e0c6bd802df8990c1c
--- /dev/null
+++ b/StreamLearn/tests/test_SAFC.py
@@ -0,0 +1,60 @@
+# -*- coding: utf-8 -*-
+"""
+Created on Fri Nov 22 20:02:07 2024
+
+@author: zhangxinyue
+"""
+
+import time
+
+import numpy as np
+import math
+
+from sklearn.preprocessing import StandardScaler
+from sklearn import model_selection
+from sklearn.metrics import *
+
+from sklearn import svm
+from sklearn.svm import SVC
+from sklearn.svm import LinearSVC
+
+import scipy.io as io
+from scipy.io import loadmat
+import os
+from joblib import dump,load
+
+from types import SimpleNamespace
+
+#Ours
+from StreamLearn.Algorithm.Algorithm_SAFC.compared_method import *
+from StreamLearn.Algorithm.Algorithm_SAFC.predict import *
+from StreamLearn.Algorithm.Algorithm_SAFC.SAFC import *
+from StreamLearn.Algorithm.Algorithm_SAFC.MetricsPred import *
+from StreamLearn.Algorithm.Algorithm_SAFC.validfunc_both import *
+from StreamLearn.Algorithm.Algorithm_SAFC.Read_to_Python import *
+from StreamLearn.Algorithm.Algorithm_SAFC.SAFC_Stream_Funcs import *
+from StreamLearn.Algorithm.Algorithm_SAFC.Class_CIFAR10_SAFC_DandSAFC_ID import *
+
+#%%
+def train_and_evaluate_stream_SAFC(args_address,runT):
+    alg_SAFC=SAFC_achieve(args_address)
+    for i in range(runT):
+        alg_SAFC.stream_fit()
+        alg_SAFC.stream_evaluate()
+    return
+    
+def main():
+    # 地址参数定义
+    data_batch_dir="Dataset//"
+    args_address=SimpleNamespace(
+        save_dir="ModelsIN",
+        PathSet=[data_batch_dir+"data_batch_1.mat",data_batch_dir+"data_batch_2.mat",data_batch_dir+"data_batch_3.mat",data_batch_dir+"data_batch_4.mat"],
+        new_path=[data_batch_dir+"data_batch_5.mat"],
+        test_path=data_batch_dir+"test_batch.mat")
+    # 运行次数定义
+    runT=3
+    # 运行
+    train_and_evaluate_stream_SAFC(args_address,runT)
+    
+if __name__ == '__main__':
+    main()
\ No newline at end of file
diff --git a/StreamLearn/tests/tests_SAFC/CIFAR10_SAFC_DandSAFC_ID_allFunc_main.py b/StreamLearn/tests/tests_SAFC/CIFAR10_SAFC_DandSAFC_ID_allFunc_main.py
deleted file mode 100644
index 4129044407289043f1d9ab7201872efa7290a7b2..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/CIFAR10_SAFC_DandSAFC_ID_allFunc_main.py
+++ /dev/null
@@ -1,279 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-import time
-
-import numpy as np
-import math
-
-from sklearn.preprocessing import StandardScaler
-from sklearn import model_selection
-from sklearn.metrics import *
-
-from compared_method import *
-from predict import *
-from SAFC import *
-from MetricsPred import *
-from validfunc_both import *
-from Read_to_Python import *
-from SAFC_Stream_Funcs import *
-
-from sklearn import svm
-from sklearn.svm import SVC
-from sklearn.svm import LinearSVC
-
-import scipy.io as io
-from scipy.io import loadmat
-import os
-from joblib import dump,load
-
-# %% Pre-define
-# 定义一个稀疏矩阵的方法
-def ind2vec(indices, num_classes):
-    indices = np.asarray(indices)
-    out = np.zeros((num_classes, np.shape(indices)[1]))
-    for i in range(num_classes):
-        index3 = np.where(i + 1 == indices)[1]
-        out[i, index3] = 1
-    return out
-
-def datareconsrtuct(X,Y):
-    #从大到小排列Y
-    sorted_index=np.argsort(Y, axis=None)
-    newY=Y[sorted_index]
-    #相应替换X的位置
-    newX=X[sorted_index]
-    return newX,newY
-
-# %%
-def SAFC_DandSAFC_ID_streamdata_all(PathSet,new_path,test_path):
-    # 存储地址
-    save_dir="ModelsIN"
-    
-    # %%Parameters and given values
-    # Metrics
-    Acc_ours1, Acc_ours2= [], []
-    AUC_ours1, AUC_ours2 = [], []
-    F1_weight_ours1, F1_weight_ours2 = [], []
-    F1_macro_ours1, F1_macro_ours2 = [], []
-    F1_micro_ours1, F1_micro_ours2 = [], []
-    
-    # Paras
-    eta = 0.1
-    alpha_set = [10 ** -2, 10 ** -1, 10 ** 0]
-    alpha = 0.001
-    beta = 0.01  
-    
-    # %% 第一阶段数据读取与模型训练
-    print('####eval####')
-    print("开始读取第一阶段数据！")
-    print('####eval####')
-    
-    X_past,Y_past=readbatchtoPython(PathSet)
-    # datareconstruction
-    X_past,Y_past=datareconsrtuct(X_past,Y_past)
-    Y_past=Y_past+1
-    
-    print('####eval####')
-    print("第一阶段数据读取完成！")
-    print('####eval####')
-    
-    # Given and transform
-    data_s1=X_past
-    label_s1_vec=Y_past
-    label_s1 = ind2vec (Y_past.T, len(np.unique(Y_past)))
-    
-    # 归一化
-    print('####eval####')
-    print("第一阶段数据归一化！")
-    print('####eval####')
-    scaler = StandardScaler()
-    data_s1 = scaler.fit_transform(data_s1)
-    print('####eval####')
-    print("第一阶段数据归一化完成！")
-    print('####eval####')
-    
-    # 第一阶段数据特征输出
-    print('####eval####')
-    print("第一阶段数据：数据量：{}，特征维度：{}，类别数：{}".format(np.shape(data_s1)[0], np.shape(data_s1)[1], len(np.unique(Y_past))))
-    print('####eval####')
-    
-    # 第一阶段模型训练与存储，用SVM训练第一阶段模型
-    print('####eval####')
-    print("begin svm1 training!")
-    print('####eval####')
-    svm1 = LinearSVC()
-    svm1.fit(data_s1.tolist(), label_s1_vec.tolist())
-    dump(svm1, save_dir+'/svm1.model')
-    print('####eval####')
-    print("end svm1 training!")
-    print('####eval####')
-    
-    # %% 第二阶段数据读取与模型训练
-    print('####eval####')
-    print("开始读取第二阶段数据！")
-    print('####eval####')
-    
-    X_new,Y_new=readtopython(new_path)
-    X_new,Y_new=datareconsrtuct(X_new,Y_new)
-    Y_new=Y_new+1
-    
-    print('####eval####')
-    print("第二阶段数据读取完成！")
-    print('####eval####')
-    
-    # Given and transform
-    data_s2=X_new
-    label_s2_vec=Y_new
-    label_s2 = ind2vec (Y_new.T, len(np.unique(Y_new)))
-    
-    # 归一化
-    print('####eval####')
-    print("第二阶段数据归一化！")
-    print('####eval####')
-    scaler = StandardScaler()
-    data_s2 = scaler.fit_transform(data_s2)
-    print('####eval####')
-    print("第二阶段数据归一化完成！")
-    print('####eval####')
-    
-    # 第二阶段数据特征输出
-    print('####eval####')
-    print("第二阶段涉及特征新增与类别新增！")
-    print('####eval####')
-    print('####eval####')
-    print("第二阶段数据：数据量：{}，特征维度：{}，类别数：{}".format(np.shape(data_s2)[0], np.shape(data_s2)[1], len(np.unique(Y_new))))
-    print('####eval####')
-    
-    # 第二阶段模型训练与存储
-    # 变体一
-    alpha_best1, beta_best1 = validfunc(np.transpose(data_s2), label_s2, alpha_set, eta, "SAFC_D",np.mat(svm1.coef_))
-    startours1 = time.time()
-    print('####eval####')
-    print("begin SAFC_D training!")
-    print('####eval####')
-    w_ours1 = SAFC_D(np.mat(svm1.coef_), np.transpose(data_s2), label_s2, alpha_best1, beta_best1, eta)
-    dump(w_ours1, save_dir+'/SAFC_D.model')
-    timeours1 = time.time() - startours1
-    print('####eval####')
-    print(timeours1)
-    print('####eval####')
-    print('####eval####')
-    print("end SAFC_D training!")
-    print('####eval####')
-    
-    # 变体二
-    alpha_best2,beta_best2=validfunc(np.transpose(data_s2),label_s2,alpha_set,eta,"SAFC_ID",np.mat(svm1.coef_))
-    startours2=time.time()
-    print('####eval####')
-    print("begin SAFC_ID training!")
-    print('####eval####')
-    w_ours2 = SAFC_ID(np.mat(svm1.coef_),np.transpose(data_s2),label_s2,alpha_best2,beta_best2,eta)
-    dump(w_ours2, save_dir+'/SAFC_ID.model')
-    timeours2=time.time()-startours2
-    print('####eval####')
-    print(timeours2)
-    print('####eval####')
-    print('####eval####')
-    print("end SAFC_ID training!")
-    print('####eval####')
-    
-    # %%测试数据读取与评估
-    print('####eval####')
-    print("开始读取测试数据！")
-    print('####eval####')
-    
-    X_test,Y_test=readtopython(test_path)
-    X_test,Y_test=datareconsrtuct(X_test,Y_test)
-    Y_test=Y_test+1
-    
-    print('####eval####')
-    print("测试数据读取完成！")
-    print('####eval####')
-    
-    # Given and transform
-    test_data=X_test
-    test_label_vec=Y_test
-    test_label = ind2vec (Y_test.T, len(np.unique(Y_test)))
-    
-    # 归一化
-    print('####eval####')
-    print("测试数据归一化！")
-    print('####eval####')
-    scaler = StandardScaler()
-    test_data = scaler.fit_transform(test_data)
-    print('####eval####')
-    print("测试数据归一化完成！")
-    print('####eval####')
-    
-    # 测试数据特征输出
-    print('####eval####')
-    print("测试数据：数据量：{}，特征维度：{}，类别数：{}".format(np.shape(test_data)[0], np.shape(test_data)[1], len(np.unique(Y_test))))
-    print('####eval####')
-    
-    # Evaluation
-    # 变体一
-    print('####eval####')
-    print("begin SAFC_D testing!")
-    print('####eval####')
-    pred1, acc_ours1, auc_ours1, f1wei_ours1, f1macro_ours1, f1micro_ours1 = Predict(w_ours1,np.transpose(test_data),test_label)
-    print('####eval####')
-    print("begin SAFC_D evaluation!")
-    print('####eval####')
-    Acc_ours1.append(acc_ours1)
-    AUC_ours1.append(auc_ours1)
-    F1_weight_ours1.append(f1wei_ours1)
-    F1_macro_ours1.append(f1macro_ours1)
-    F1_micro_ours1.append(f1micro_ours1)
-    
-    # 变体二
-    print('####eval####')
-    print("begin SAFC_ID testing!")
-    print('####eval####')
-    pred2,acc_ours2,auc_ours2,f1wei_ours2,f1macro_ours2,f1micro_ours2 = Predict(w_ours2,np.transpose(test_data),test_label)
-    print('####eval####')
-    print("begin SAFC_ID evaluation!")
-    print('####eval####')
-    Acc_ours2.append(acc_ours2)
-    AUC_ours2.append(auc_ours2)
-    F1_weight_ours2.append(f1wei_ours2)
-    F1_macro_ours2.append(f1macro_ours2)
-    F1_micro_ours2.append(f1micro_ours2)
-    
-    # 结果输出
-    print('####eval####')
-    print("预测与评估完成，输出评估结果！")
-    print('####eval####')
-    
-    meanAcc_ours1 = np.mean(Acc_ours1)
-    meanAcc_ours2 = np.mean(Acc_ours2)
-    meanAuc_ours1 = np.mean(AUC_ours1)
-    meanAuc_ours2 = np.mean(AUC_ours2)
-    meanF1_macro_ours1 = np.mean(F1_macro_ours1)
-    meanF1_macro_ours2 = np.mean(F1_macro_ours2)
-    meanF1_weight_ours1 = np.mean(F1_weight_ours1)
-    meanF1_weight_ours2 = np.mean(F1_weight_ours2)
-    meanF1_micro_ours1 = np.mean(F1_micro_ours1)
-    meanF1_micro_ours2 = np.mean(F1_micro_ours2)
-    
-    print('####eval####')
-    print(
-        'meanAcc_ours1{:.4f},meanAcc_ours2{:.4f},meanAuc_ours1{:.4f},meanAuc_ours2{:.4f},meanF1_macro_ours1{:.4f},meanF1_macro_ours2{:.4f},meanF1_weight_ours1{:.4f},meanF1_weight_ours2{:.4f},meanF1_micro_ours1{:.4f},meanF1_micro_ours2{:.4f}'
-        .format(meanAcc_ours1,meanAcc_ours2, meanAuc_ours1,meanAuc_ours2,meanF1_macro_ours1,meanF1_macro_ours2,meanF1_weight_ours1,meanF1_weight_ours2,meanF1_micro_ours1,meanF1_micro_ours2))
-    print('####eval####')
-    print('Finished!')
-
-    return
-# %% 主程序调用区
-if __name__ == "__main__":
-    # 读取地址
-    data_batch_dir="Dataset//"
-    PathSet=[data_batch_dir+"data_batch_1.mat",data_batch_dir+"data_batch_2.mat",data_batch_dir+"data_batch_3.mat",data_batch_dir+"data_batch_4.mat"]
-    new_path=data_batch_dir+"data_batch_5.mat"
-    test_path=data_batch_dir+"test_batch.mat"
-    
-    SAFC_DandSAFC_ID_streamdata_all(PathSet,new_path,test_path)
\ No newline at end of file
diff --git a/StreamLearn/tests/tests_SAFC/CIFAR10_SAFC_DandSAFC_ID_useFunc_main.py b/StreamLearn/tests/tests_SAFC/CIFAR10_SAFC_DandSAFC_ID_useFunc_main.py
deleted file mode 100644
index 6f2ee543d8c0fa3a6d7572c3be5268bbae1b7c55..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/CIFAR10_SAFC_DandSAFC_ID_useFunc_main.py
+++ /dev/null
@@ -1,77 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-import time
-
-import numpy as np
-import math
-
-from sklearn.preprocessing import StandardScaler
-from sklearn import model_selection
-from sklearn.metrics import *
-
-from compared_method import *
-from predict import *
-from SAFC import *
-from MetricsPred import *
-from validfunc_both import *
-from Read_to_Python import *
-from SAFC_Stream_Funcs import *
-
-from sklearn import svm
-from sklearn.svm import SVC
-from sklearn.svm import LinearSVC
-
-import scipy.io as io
-from scipy.io import loadmat
-import os
-from joblib import dump,load
-
-# %% 第一阶段、第二阶段、测试main函数
-# 存储地址
-save_dir="ModelsIN"
-
-# 读取地址
-data_batch_dir="Dataset//"
-PathSet=[data_batch_dir+"data_batch_1.mat",data_batch_dir+"data_batch_2.mat",data_batch_dir+"data_batch_3.mat",data_batch_dir+"data_batch_4.mat"]
-new_path=data_batch_dir+"data_batch_5.mat"
-test_path=data_batch_dir+"test_batch.mat"
-
-# 第一阶段数据读取与训练
-SAFC_Stage1(PathSet,save_dir)
-# 第一阶段模型提取
-print('####eval####')
-print("Read stage1 model!")
-print('####eval####')
-stage1model=load(save_dir+'/svm1.model')
-print('####eval####')
-print("Have read stage1 model!")
-print('####eval####')
-
-# 第二阶段数据读取与训练
-SAFC_Stage2(new_path,stage1model,save_dir)
-# 第二阶段模型提取
-# 变体一
-print('####eval####')
-print("Read stage2-SAFC_D model!")
-print('####eval####')
-stage2model1=load(save_dir+'/SAFC_D.model')
-print('####eval####')
-print("Have read stage2-SAFC_D model!")
-print('####eval####')
-# 变体二
-print('####eval####')
-print("Read stage2-SAFC_D model!")
-print('####eval####')
-stage2model2=load(save_dir+'/SAFC_ID.model')
-print('####eval####')
-print("Have read stage2-SAFC_D model!")
-print('####eval####')
-
-# 测试数据读取与评估
-SAFC_test(test_path,stage2model1,stage2model2)
-
diff --git a/StreamLearn/tests/tests_SAFC/Dataset/dataset_information b/StreamLearn/tests/tests_SAFC/Dataset/dataset_information
deleted file mode 100644
index d04db552d4f06fe538020dcfb5ea4f1e60365274..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/Dataset/dataset_information
+++ /dev/null
@@ -1,4 +0,0 @@
-数据集
-通过百度网盘分享的文件：SAFC_datasets_CIFAR10.zip
-链接：https://pan.baidu.com/s/1xtZjSxIIEMnUwoM7VXCzkQ 
-提取码：nudt
\ No newline at end of file
diff --git a/StreamLearn/tests/tests_SAFC/MetricsPred.py b/StreamLearn/tests/tests_SAFC/MetricsPred.py
deleted file mode 100644
index 84c63597090711c9be48bfa3edeb846ad70d5cb5..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/MetricsPred.py
+++ /dev/null
@@ -1,19 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-import numpy as np
-from sklearn.metrics import *
-
-# %%
-def MetricsPred(test_label,test_label_mat,pred,prob):
-    
-    acc = np.mean(test_label.flatten()== pred.flatten())
-    AUC=roc_auc_score(test_label_mat.tolist(),prob.tolist())
-    F1_weight=f1_score(test_label.tolist(),pred.tolist(), average='weighted')
-    F1_macro=f1_score(test_label.tolist(),pred.tolist(), average='macro')
-    F1_micro=f1_score(test_label.tolist(),pred.tolist(), average='micro')
-    return acc,AUC,F1_weight,F1_macro,F1_micro
\ No newline at end of file
diff --git a/StreamLearn/tests/tests_SAFC/Read_to_Python.py b/StreamLearn/tests/tests_SAFC/Read_to_Python.py
deleted file mode 100644
index c8486ba84ad3b782e7aa885fcafb5c360f29cfae..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/Read_to_Python.py
+++ /dev/null
@@ -1,41 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-import numpy as np
-from scipy.io import loadmat
-
-# %%
-# tips
-# import os
-# print(f"当前工作目录: {os.getcwd()}")
-
-# 数据接口
-def readtopython(path):
-    data_part=loadmat(path)
-    X_part=data_part["data"].astype(np.float64)
-    Y_part=data_part["labels"].astype(np.float64)
-    return X_part,Y_part
-    
-# Batch数据接口
-def readbatchtoPython(Path):
-    for path_index in range(len(Path)):
-        path=Path[path_index]
-        X_part,Y_part=readtopython(path)
-        if path_index==0:
-            X_past_original=X_part
-            Y_past_original=Y_part
-        if path_index!=0:
-            X_past_original=np.vstack((X_past_original,X_part))
-            Y_past_original=np.vstack((Y_past_original,Y_part))
-    # 第一阶段的数据没有No.9，即第10类数据
-    Position_past=np.where(Y_past_original!=9)[0]
-    Y_past=Y_past_original[Position_past]
-    # 特征不含B通道的1024维特征
-    X_past=X_past_original[Position_past,0:2048]
-    return X_past,Y_past
-
-
diff --git a/StreamLearn/tests/tests_SAFC/SAFC.py b/StreamLearn/tests/tests_SAFC/SAFC.py
deleted file mode 100644
index dbf74f489e854d7f5e4a8edc0dd816903c48319e..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/SAFC.py
+++ /dev/null
@@ -1,165 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-import numpy as np
-import math
-
-#%% slove Memory Error key
-#redefine X_21@D:
-def AxdiagB(A,B):
-    #A=d*N,B=1*N
-    BL=B.tolist()[0]
-    C=[]
-    for i in range(np.shape(A)[0]):
-        C_row=[]
-        for j in range(np.shape(A)[1]):
-            ele=BL[j]*A[i,j]
-            C_row.append(ele)
-        C.append(C_row)
-    C_mat=np.mat(C)
-    return C_mat
-#%%
-# 函数1、SAFC_D
-def SAFC_D(W1,train_data,train_label,alpha,beta,eta):
-##%% Input:
-#% W1: c*d1, a model learned in previous stage, and it will be reused in 
-#%     current stage. W1 can be learned by using 'mysoftmax' function.
-#%     c and d1 are the number of classes and the dimension of data in
-#%     previous stage, respectively.
-#% train_data: d2*n, where d2 is the dimension of data in current stage, d2 > d1.
-#%             n is the number of data in current stage.
-#% train_label: (c+1)*n, each column is a one-hot vector. 
-#% alpha and beta: the hyperparameters in the model, alpha > 0, beta >0.
-#% eta: the step size in the gradient descent algorithm.
-##%% Output:
-#%  softmaxModel: (c+1)*d, the learned classifier.
-##%% Our proposed method
-    cost_old = 0 #%set old cost and new cost value
-    cost = 1
-    object_value = [] #% record the cost at each update iteration
-    count = 0 #% count the running number    
-    nCla1,nFea1 = np.shape(W1)
-    nCla2,nSam2 = np.shape(train_label)
-    nFea2 = np.shape(train_data)[0]
-    X_21 = train_data[0:nFea1,:]
-    X_22 = train_data[nFea1:,:]
-    W2 = 0.005*np.ones((nCla2,nFea2))   #%  Initialise classifier W
-    #%W2= mysoftmax(data_s2,label_s2,alpha_set, eta)[0]
-    W21 = W2[:,0:nFea1]
-    W22 = W2[:,nFea1:]
-    W21_tilde = W21[0:nCla1,:]
-    W21_hat = W21[nCla1,:]
-    W22_tilde = W22[0:nCla1,:]
-    W22_hat = W22[nCla1,:]
-    loop_max=3000
-    while (abs(cost_old - cost) > 10^-6 ) and ( count < loop_max ):
-        #%( abs(cost_old - cost) > 0.0001*cost ) &&
-         cost_old = cost
-         count=count+1
-         #M = bsxfun(@minus,W2*train_data,max(W2*train_data, [], 1))
-         # 计算 W2_train_data 和 col_max
-         W2_train_data = np.dot(W2, train_data)
-         col_max = np.max(W2_train_data, axis=0)
-         M = W2_train_data - col_max
-         M = np.exp(M)
-         #p = bsxfun(@rdivide, M, sum(M));
-         col_sum = np.sum(M, axis=0)
-         divisor = np.tile(col_sum, (M.shape[0], 1))
-         p= np.divide(M, divisor)
-         cost = -1/nSam2 * train_label.flatten() @ np.log(p.flatten()).T + alpha * np.sum(np.square(W2))+beta *np.sum(np.square(W21_tilde.flatten()-W1.flatten()))
-         W21_tilde_grad = -1/nSam2 * (train_label[0:nCla1,:] - p[0:nCla1,:]) @ X_21.T + 2*alpha*W21_tilde + 2*beta * (W21_tilde-W1)
-         W21_hat_grad =  -1/nSam2 * (train_label[nCla1,:] - p[nCla1,:]) @ X_21.T + 2*alpha*W21_hat
-         W22_grad = -1/nSam2 * (train_label - p) @ X_22.T + 2*alpha * W22
-         W21_tilde = W21_tilde - eta*W21_tilde_grad
-         W21_hat = W21_hat - eta*W21_hat_grad
-         W22 = W22 - eta*W22_grad
-         W21=np.vstack(((W21_tilde),W21_hat))
-         W2=np.hstack(((W21),W22))
-         if object_value==[]:
-            np.mat(object_value.append(cost))
-         else:
-            object_value=np.vstack(((object_value),cost))
-    softmaxModel = W2
-    return softmaxModel
-
-#%%
-# 函数2、SAFC_ID函数
-def SAFC_ID(W1, train_data, train_label, alpha,beta, eta):
-##%% Input:
-#% W1: c*d1, a model learned in previous stage, and it will be reused in 
-#%     current stage. W1 can be learned by using 'mysoftmax' function.
-#%     c and d1 are the number of classes and the dimension of data in
-#%     previous stage, respectively.
-#% train_data: d2*n, where d2 is the dimension of data in current stage, d2 > d1.
-#%             n is the number of data in current stage.
-#% train_label: (c+1)*n, each column is a one-hot vector. 
-#% alpha and beta: the hyperparameters in the model, alpha > 0, beta >0.
-#% eta: the step size in the gradient descent algorithm.
-##%% Output:
-#%  softmaxModel: (c+1)*d, the learned classifier .
-##%% Our proposed method
-#%set old cost and new cost value
-    cost_old = 0 #%set old cost and new cost value
-    cost = 1
-    object_value = [] #% record the cost at each update iteration
-    count = 0 #% count the running number    
-    nCla1,nFea1 = np.shape(W1)
-    nCla2,nSam2 = np.shape(train_label)
-    nFea2 = np.shape(train_data)[0]
-    X_21 = train_data[0:nFea1,:]
-    X_22 = train_data[nFea1:,:]
-    W2 = 0.005*np.ones((nCla2,nFea2))   #%  Initialise classifier W
-    #%W2= mysoftmax(data_s2,label_s2,alpha_set, eta)[0]
-    W21 = W2[:,0:nFea1]
-    W22 = W2[:,nFea1:]
-    W21_tilde = W21[0:nCla1,:]
-    W21_hat = W21[nCla1,:]
-    W22_tilde = W22[0:nCla1,:]
-    W22_hat = W22[nCla1,:]
-    epsilon = 0.00001
-   # % epsilon = 1;
-    loop_max=3000
-    while (abs(cost_old - cost) > 10^-6 ) and ( count < loop_max ):
-        #%( abs(cost_old - cost) > 0.0001*cost ) &&
-         cost_old = cost
-         count=count+1
-         #M = bsxfun(@minus,W2*train_data,max(W2*train_data, [], 1))
-         # 计算 W2_train_data 和 col_max
-         W2_train_data = np.dot(W2, train_data)
-         col_max = np.max(W2_train_data, axis=0)
-         M = W2_train_data - col_max
-         M = np.exp(M)
-         #p = bsxfun(@rdivide, M, sum(M));
-         col_sum = np.sum(M, axis=0)
-         divisor = np.tile(col_sum, (M.shape[0], 1))
-         p= np.divide(M, divisor)
-         reg = (W21_tilde@X_21-W1@X_21).T
-         norm_column = []
-         for i in range(nSam2): 
-             norm_column.append(np.linalg.norm(reg[i,:]))
-         norm_column=np.mat(norm_column)    
-         MM = 1.0/(norm_column+epsilon)
-         #D = np.diag(MM.tolist()[0])#%     D = eye(nSam2)Memory Error
-         
-         cost = -1/nSam2 * train_label.flatten() @ np.log(p.flatten()).T + alpha * np.sum(np.square(W2))+beta *np.sum(norm_column[0:])
-         W21_tilde_grad = -1/nSam2 * (train_label[0:nCla1,:] - p[0:nCla1,:]) @ X_21.T + 2*alpha*W21_tilde + beta *(AxdiagB(X_21,MM)@(W21_tilde@X_21-W1@X_21).T).T
-         
-         W21_hat_grad =  -1/nSam2 * (train_label[nCla1,:] - p[nCla1,:]) @ X_21.T + 2*alpha*W21_hat
-         W22_grad = -1/nSam2 * (train_label - p) @ X_22.T + 2*alpha * W22
-         
-         W21_tilde = W21_tilde - eta*W21_tilde_grad
-         W21_hat = W21_hat - eta*W21_hat_grad
-         W22 = W22 - eta*W22_grad
-         W21=np.vstack(((W21_tilde),W21_hat))
-         W2=np.hstack(((W21),W22))
-         if object_value==[]:
-            np.mat(object_value.append(cost))
-         else:
-            object_value=np.vstack(((object_value),cost))
-    softmaxModel = W2
-    
-    return softmaxModel
\ No newline at end of file
diff --git a/StreamLearn/tests/tests_SAFC/SAFC_Stream_Funcs.py b/StreamLearn/tests/tests_SAFC/SAFC_Stream_Funcs.py
deleted file mode 100644
index 343167d277f31ab48bdc74f909b7202c146a2858..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/SAFC_Stream_Funcs.py
+++ /dev/null
@@ -1,273 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-import time
-
-import numpy as np
-import math
-
-from sklearn.preprocessing import StandardScaler
-from sklearn import model_selection
-from sklearn.metrics import *
-
-from compared_method import *
-from predict import *
-from SAFC import *
-from MetricsPred import *
-from validfunc_both import *
-from Read_to_Python import *
-from SAFC_Stream_Funcs import *
-
-from sklearn import svm
-from sklearn.svm import SVC
-from sklearn.svm import LinearSVC
-
-import scipy.io as io
-from scipy.io import loadmat
-import os
-from joblib import dump,load
-
-# %% Adress and Pre-define
-# 定义一个稀疏矩阵的方法
-def ind2vec(indices, num_classes):
-    indices = np.asarray(indices)
-    out = np.zeros((num_classes, np.shape(indices)[1]))
-    for i in range(num_classes):
-        index3 = np.where(i + 1 == indices)[1]
-        out[i, index3] = 1
-    return out
-
-def datareconsrtuct(X,Y):
-    #从大到小排列Y
-    sorted_index=np.argsort(Y, axis=None)
-    newY=Y[sorted_index]
-    #相应替换X的位置
-    newX=X[sorted_index]
-    return newX,newY
-
-# %% 定义第一阶段的读取训练与存储
-def SAFC_Stage1(PathSet,save_dir):
-    # 第一阶段数据读取与模型训练
-    print('####eval####')
-    print("开始读取第一阶段数据！")
-    print('####eval####')
-    
-    X_past,Y_past=readbatchtoPython(PathSet)
-    X_past,Y_past=datareconsrtuct(X_past,Y_past)
-    Y_past=Y_past+1
-
-    print('####eval####')
-    print("第一阶段数据读取完成！")
-    print('####eval####')
-
-    # Given and transform
-    data_s1=X_past
-    label_s1_vec=Y_past
-    label_s1 = ind2vec (Y_past.T, len(np.unique(Y_past)))
-
-    # 归一化
-    print('####eval####')
-    print("第一阶段数据归一化！")
-    print('####eval####')
-    scaler = StandardScaler()
-    data_s1 = scaler.fit_transform(data_s1)
-    print('####eval####')
-    print("第一阶段数据归一化完成！")
-    print('####eval####')
-
-    # 第一阶段数据特征输出
-    print('####eval####')
-    print("第一阶段数据：数据量：{}，特征维度：{}，类别数：{}".format(np.shape(data_s1)[0], np.shape(data_s1)[1], len(np.unique(Y_past))))
-    print('####eval####')
-
-    # 第一阶段模型训练与存储，用SVM训练第一阶段模型
-    print('####eval####')
-    print("begin svm1 training!")
-    print('####eval####')
-    svm1 = LinearSVC()
-    svm1.fit(data_s1.tolist(), label_s1_vec.tolist())
-    dump(svm1, save_dir+'/svm1.model')
-    print('####eval####')
-    print("end svm1 training!")
-    print('####eval####')
-    
-    return
-
-# %% 定义第二阶段的读取训练与存储
-def SAFC_Stage2(new_path,svm1,save_dir):
-    # Paras
-    eta = 0.1
-    alpha_set = [10 ** -2, 10 ** -1, 10 ** 0]
-    alpha = 0.001
-    beta = 0.01  
-    
-    # 第二阶段数据读取与模型训练
-    print('####eval####')
-    print("开始读取第二阶段数据！")
-    print('####eval####')
-
-    X_new,Y_new=readtopython(new_path)
-    X_new,Y_new=datareconsrtuct(X_new,Y_new)
-    Y_new=Y_new+1
-
-    print('####eval####')
-    print("第二阶段数据读取完成！")
-    print('####eval####')
-
-    # Given and transform
-    data_s2=X_new
-    label_s2_vec=Y_new
-    label_s2 = ind2vec (Y_new.T, len(np.unique(Y_new)))
-
-    # 归一化
-    print('####eval####')
-    print("第二阶段数据归一化！")
-    print('####eval####')
-    scaler = StandardScaler()
-    data_s2 = scaler.fit_transform(data_s2)
-    print('####eval####')
-    print("第二阶段数据归一化完成！")
-    print('####eval####')
-
-    # 第二阶段数据特征输出
-    print('####eval####')
-    print("第二阶段涉及特征新增与类别新增！")
-    print('####eval####')
-    print('####eval####')
-    print("第二阶段数据：数据量：{}，特征维度：{}，类别数：{}".format(np.shape(data_s2)[0], np.shape(data_s2)[1], len(np.unique(Y_new))))
-    print('####eval####')
-
-    # 第二阶段模型训练与存储
-    # 变体一
-    alpha_best1, beta_best1 = validfunc(np.transpose(data_s2), label_s2, alpha_set, eta, "SAFC_D",np.mat(svm1.coef_))
-    startours1 = time.time()
-    print('####eval####')
-    print("begin SAFC_D training!")
-    print('####eval####')
-    w_ours1 = SAFC_D(np.mat(svm1.coef_), np.transpose(data_s2), label_s2, alpha_best1, beta_best1, eta)
-    dump(w_ours1, save_dir+'/SAFC_D.model')
-    timeours1 = time.time() - startours1
-    print('####eval####')
-    print(timeours1)
-    print('####eval####')
-    print('####eval####')
-    print("end SAFC_D training!")
-    print('####eval####')
-
-    # 变体二
-    alpha_best2,beta_best2=validfunc(np.transpose(data_s2),label_s2,alpha_set,eta,"SAFC_ID",np.mat(svm1.coef_))
-    startours2=time.time()
-    print('####eval####')
-    print("begin SAFC_ID training!")
-    print('####eval####')
-    w_ours2 = SAFC_ID(np.mat(svm1.coef_),np.transpose(data_s2),label_s2,alpha_best2,beta_best2,eta)
-    dump(w_ours2, save_dir+'/SAFC_ID.model')
-    timeours2=time.time()-startours2
-    print('####eval####')
-    print(timeours2)
-    print('####eval####')
-    print('####eval####')
-    print("end SAFC_ID training!")
-    print('####eval####')
-    
-    return
-
-# %% 定义测试数据读取与评估
-def SAFC_test(test_path,w_ours1,w_ours2):
-    # Metrics
-    Acc_ours1, Acc_ours2= [], []
-    AUC_ours1, AUC_ours2 = [], []
-    F1_weight_ours1, F1_weight_ours2 = [], []
-    F1_macro_ours1, F1_macro_ours2 = [], []
-    F1_micro_ours1, F1_micro_ours2 = [], []
-    
-    # 测试数据读取与评估
-    print('####eval####')
-    print("开始读取测试数据！")
-    print('####eval####')
-
-    X_test,Y_test=readtopython(test_path)
-    X_test,Y_test=datareconsrtuct(X_test,Y_test)
-    Y_test=Y_test+1
-
-    print('####eval####')
-    print("测试数据读取完成！")
-    print('####eval####')
-
-    # Given and transform
-    test_data=X_test
-    test_label_vec=Y_test
-    test_label = ind2vec (Y_test.T, len(np.unique(Y_test)))
-
-    # 归一化
-    print('####eval####')
-    print("测试数据归一化！")
-    print('####eval####')
-    scaler = StandardScaler()
-    test_data = scaler.fit_transform(test_data)
-    print('####eval####')
-    print("测试数据归一化完成！")
-    print('####eval####')
-
-    # 测试数据特征输出
-    print('####eval####')
-    print("测试数据：数据量：{}，特征维度：{}，类别数：{}".format(np.shape(test_data)[0], np.shape(test_data)[1], len(np.unique(Y_test))))
-    print('####eval####')
-
-    # Evaluation
-    # 变体一
-    print('####eval####')
-    print("begin SAFC_D testing!")
-    print('####eval####')
-    pred1, acc_ours1, auc_ours1, f1wei_ours1, f1macro_ours1, f1micro_ours1 = Predict(w_ours1,np.transpose(test_data),test_label)
-    print('####eval####')
-    print("begin SAFC_D evaluation!")
-    print('####eval####')
-    Acc_ours1.append(acc_ours1)
-    AUC_ours1.append(auc_ours1)
-    F1_weight_ours1.append(f1wei_ours1)
-    F1_macro_ours1.append(f1macro_ours1)
-    F1_micro_ours1.append(f1micro_ours1)
-
-    # 变体二
-    print('####eval####')
-    print("begin SAFC_ID testing!")
-    print('####eval####')
-    pred2,acc_ours2,auc_ours2,f1wei_ours2,f1macro_ours2,f1micro_ours2 = Predict(w_ours2,np.transpose(test_data),test_label)
-    print('####eval####')
-    print("begin SAFC_ID evaluation!")
-    print('####eval####')
-    Acc_ours2.append(acc_ours2)
-    AUC_ours2.append(auc_ours2)
-    F1_weight_ours2.append(f1wei_ours2)
-    F1_macro_ours2.append(f1macro_ours2)
-    F1_micro_ours2.append(f1micro_ours2)
-
-    # 结果输出
-    print('####eval####')
-    print("预测与评估完成，输出评估结果！")
-    print('####eval####')
-
-    meanAcc_ours1 = np.mean(Acc_ours1)
-    meanAcc_ours2 = np.mean(Acc_ours2)
-    meanAuc_ours1 = np.mean(AUC_ours1)
-    meanAuc_ours2 = np.mean(AUC_ours2)
-    meanF1_macro_ours1 = np.mean(F1_macro_ours1)
-    meanF1_macro_ours2 = np.mean(F1_macro_ours2)
-    meanF1_weight_ours1 = np.mean(F1_weight_ours1)
-    meanF1_weight_ours2 = np.mean(F1_weight_ours2)
-    meanF1_micro_ours1 = np.mean(F1_micro_ours1)
-    meanF1_micro_ours2 = np.mean(F1_micro_ours2)
-
-    print('####eval####')
-    print(
-        'meanAcc_ours1{:.4f},meanAcc_ours2{:.4f},meanAuc_ours1{:.4f},meanAuc_ours2{:.4f},meanF1_macro_ours1{:.4f},meanF1_macro_ours2{:.4f},meanF1_weight_ours1{:.4f},meanF1_weight_ours2{:.4f},meanF1_micro_ours1{:.4f},meanF1_micro_ours2{:.4f}'
-        .format(meanAcc_ours1,meanAcc_ours2, meanAuc_ours1,meanAuc_ours2,meanF1_macro_ours1,meanF1_macro_ours2,meanF1_weight_ours1,meanF1_weight_ours2,meanF1_micro_ours1,meanF1_micro_ours2))
-    print('####eval####')
-    print('Finished!')
-    
-    return
\ No newline at end of file
diff --git a/StreamLearn/tests/tests_SAFC/compared_method.py b/StreamLearn/tests/tests_SAFC/compared_method.py
deleted file mode 100644
index 613c1c2164f03b41f5c47f835da3894fece94d25..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/compared_method.py
+++ /dev/null
@@ -1,242 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-import numpy as np
-import math
-from sklearn import model_selection
-
-from predict import * 
-
-#%% slove Memory Error key
-#redefine X_21@D:
-def AxdiagB(A,B):
-    #A=d*N,B=1*N
-    BL=B.tolist()[0]
-    C=[]
-    for i in range(np.shape(A)[0]):
-        C_row=[]
-        for j in range(np.shape(A)[1]):
-            ele=BL[j]*A[i,j]
-            C_row.append(ele)
-        C.append(C_row)
-    C_mat=np.mat(C)
-    return C_mat
-
-# %%
-# 函数1-1、mysoftmax函数中涉及到的find_best_w函数
-def find_best_w(train_data,train_label, alpha, eta):
-    [n_Fea,n_Sam] = np.shape(train_data)
-    nCla = np.shape(train_label)[0]
-    W = 0.005*np.ones((nCla,n_Fea))
-    count=0   #% count the running number
-    #%set old cost and new cost value
-    cost_old=0
-    cost=1
-    object_value=[]  #% record the cost at each update iteration
-    loop_max=3000
-    while (abs(cost_old - cost) > 10^-6)and (count < loop_max ):
-        #%( abs(cost_old - cost) > 0.0001*cost ) &&
-        cost_old = cost
-        count=count+1
-        #M = bsxfun(@minus,W*train_data,max(W*train_data, [], 1));
-        # 计算 W_train_data 和 col_max
-        W_train_data = np.dot(W, train_data)
-        col_max = np.max(W_train_data, axis=0)
-        M = W_train_data - col_max
-        M = np.exp(M)
-        #p = bsxfun(@rdivide, M, sum(M));
-        col_sum = np.sum(M, axis=0)
-        divisor = np.tile(col_sum, (M.shape[0], 1))
-        p= np.divide(M, divisor)
-        #.flatten()张成向量
-        cost = -1/n_Sam * train_label.flatten() @ np.log(p.flatten()).T + alpha * np.sum(np.square(W))
-        W_grad = -1/n_Sam * (train_label - p) @ train_data.T + 2*alpha * W
-        W = W - eta*W_grad
-        if object_value==[]:
-            object_value = np.vstack((np.tile(object_value, (cost.shape[1],1)).T, cost))
-        else:
-            object_value=np.vstack(((object_value),cost))
-    W1=W
-    return W1
-
-#%%
-# 函数1、mysoftmax函数
-def mysoftmax( train_data,train_label,alpha_set,eta):
-##%% Input:
-#% train_data: d*n, where d is the dimension of data.
-#%             n is the number of data.
-#% train_label: c*n, each column is a one-hot vector. 
-#% alpha_set: the hyperparameters set, which is used for cross-validation.
-#% eta: the step size in the gradient descent algorithm.
-##%% Output:
-#%  softmaxModel: c*d, the learned classifier.
-#%  alpha_best: the best hyperparameter determined by cross-validation.
-##%% cross validation
-    if len(alpha_set)>1:
-        alpha_num = len(alpha_set)
-        n = np.shape(train_data)[1]
-        k_fold = 5
-        #Indices = crossvalind('Kfold', n, k_fold)
-        kf=model_selection.KFold(n_splits=5,shuffle=False,random_state=None)
-        acc_statistic = np.zeros((alpha_num,1))
-        for train_index, test_index in kf.split(train_data.T):
-            #print("TRAIN:", train_index, "TEST:", test_index)
-            x_train, x_test = train_data.T[train_index].T, train_data.T[test_index].T
-            y_train, y_test = train_label.T[train_index].T, train_label.T[test_index].T  
-            for i in range(1,alpha_num+1):
-                alpha = alpha_set[i-1]
-                w_i = find_best_w(x_train,y_train,alpha,eta)
-                acc_i= Predict(w_i,x_test, y_test)[1]
-                acc_statistic[i-1]= acc_statistic[i-1]+acc_i
-        alpha_position = np.argmax(acc_statistic)
-        alpha_best = alpha_set[alpha_position]
-        softmaxModel = find_best_w( train_data,train_label, alpha_best, eta)
-    else:
-        alpha_best = alpha_set[0]
-        #由于w_s2当中使用mysoftmax函数时候，传入的是浮点数alpha列表化之后的形式，所以这里要取一下浮点数
-        softmaxModel = find_best_w( train_data,train_label, alpha_best, eta)        
-    return softmaxModel,alpha_best
-
-#%% 
-# 函数2、对比方法文件夹中的Left1函数
-def Left1(W1,train_data,train_label,alpha,beta,eta):
-##%% Input:
-#% W1: c*d1, a model learned in previous stage, and it will be reused in 
-#%     current stage. W1 can be learned by using 'mysoftmax' function.
-#%     c and d1 are the number of classes and the dimension of data in
-#%     previous stage, respectively.
-#% train_data: d2*n, where d2 is the dimension of data in current stage, d2 > d1.
-#%             n is the number of data in current stage.
-#% train_label: (c+1)*n, each column is a one-hot vector. 
-#% alpha and beta: the hyperparameters in the model, alpha > 0, beta >0.
-#% eta: the step size in the gradient descent algorithm.
-##%% Output:
-#%  softmaxModel: (c+1)*d, the learned classifier.
-##%% Our proposed method
-    cost_old = 0 #%set old cost and new cost value
-    cost = 1
-    object_value = [] #% record the cost at each update iteration
-    count = 0 #% count the running number    
-    nCla1,nFea1 = np.shape(W1)
-    nCla2,nSam2 = np.shape(train_label)
-    nFea2 = np.shape(train_data)[0]
-    X_21 = train_data[0:nFea1,:]
-    X_22 = train_data[nFea1:,:]
-    W2 = 0.005*np.ones((nCla2,nFea2))   #%  Initialise classifier W
-    #%W2= mysoftmax(data_s2,label_s2,alpha_set, eta)[0]
-    W21 = W2[:,0:nFea1]
-    W22 = W2[:,nFea1:]
-    W21_tilde = W21[0:nCla1,:]
-    W21_hat = W21[nCla1,:]
-    W22_tilde = W22[0:nCla1,:]
-    W22_hat = W22[nCla1,:]
-    loop_max=3000
-    while (abs(cost_old - cost) > 10^-6 ) and ( count < loop_max ):
-        #%( abs(cost_old - cost) > 0.0001*cost ) &&
-         cost_old = cost
-         count=count+1
-         #M = bsxfun(@minus,W21*X_21,max(W21*X_21, [], 1))
-         # 计算 W2_X_21 和 col_max
-         W21_X_21= np.dot(W21, X_21)
-         col_max = np.max(W21_X_21, axis=0)
-         M = W21_X_21 - col_max
-         M = np.exp(M)
-         #p = bsxfun(@rdivide, M, sum(M));
-         col_sum = np.sum(M, axis=0)
-         divisor = np.tile(col_sum, (M.shape[0], 1))
-         p= np.divide(M, divisor)
-         cost = -1/nSam2 * train_label.flatten() @ np.log(p.flatten()).T + alpha * np.sum(np.square(W21.flatten()))+beta *np.sum(np.square(W21_tilde.flatten()-W1.flatten()))
-         W21_tilde_grad = -1/nSam2 * (train_label[0:nCla1,:] - p[0:nCla1,:]) @ X_21.T + 2*alpha*W21_tilde + 2*beta * (W21_tilde-W1)
-         W21_hat_grad =  -1/nSam2 * (train_label[nCla1,:] - p[nCla1,:]) @ X_21.T + 2*alpha*W21_hat
-         
-         W21_tilde = W21_tilde - eta*W21_tilde_grad
-         W21_hat = W21_hat - eta*W21_hat_grad
-         
-         W21=np.vstack(((W21_tilde),W21_hat))
-       
-         if object_value==[]:
-            object_value = np.vstack((np.tile(object_value, (cost.shape[1],1)).T, cost))
-         else:
-            object_value=np.vstack(((object_value),cost))
-            
-    softmaxModel = W21
-    return softmaxModel
-
-#%%
-# 函数3、对比方法文件夹中的Left2函数
-def Left2(W1, train_data, train_label, alpha,beta, eta):
-##%% Input:
-#% W1: c*d1, a model learned in previous stage, and it will be reused in 
-#%     current stage. W1 can be learned by using 'mysoftmax' function.
-#%     c and d1 are the number of classes and the dimension of data in
-#%     previous stage, respectively.
-#% train_data: d2*n, where d2 is the dimension of data in current stage, d2 > d1.
-#%             n is the number of data in current stage.
-#% train_label: (c+1)*n, each column is a one-hot vector. 
-#% alpha and beta: the hyperparameters in the model, alpha > 0, beta >0.
-#% eta: the step size in the gradient descent algorithm.
-##%% Output:
-#%  softmaxModel: (c+1)*d, the learned classifier .
-##%% Our proposed method
-#%set old cost and new cost value
-    cost_old = 0 #%set old cost and new cost value
-    cost = 1
-    object_value = [] #% record the cost at each update iteration
-    count = 0 #% count the running number    
-    nCla1,nFea1 = np.shape(W1)
-    nCla2,nSam2 = np.shape(train_label)
-    nFea2 = np.shape(train_data)[0]
-    X_21 = train_data[0:nFea1,:]
-    X_22 = train_data[nFea1:,:]
-    W2 = 0.005*np.ones((nCla2,nFea2))   #%  Initialise classifier W
-    #%W2= mysoftmax(data_s2,label_s2,alpha_set, eta)[0]
-    W21 = W2[:,0:nFea1]
-    W22 = W2[:,nFea1:]
-    W21_tilde = W21[0:nCla1,:]
-    W21_hat = W21[nCla1,:]
-    W22_tilde = W22[0:nCla1,:]
-    W22_hat = W22[nCla1,:]
-    epsilon = 0.00001
-   # % epsilon = 1;
-    loop_max=3000
-    while (abs(cost_old - cost) > 10^-6 ) and ( count < loop_max ):
-        #%( abs(cost_old - cost) > 0.0001*cost ) &&
-         cost_old = cost
-         count=count+1
-         #M = bsxfun(@minus,W21*X_21,max(W21*X_21, [], 1))
-         # 计算 W2_X_21 和 col_max
-         W21_X_21= np.dot(W21, X_21)
-         col_max = np.max(W21_X_21, axis=0)
-         M = W21_X_21 - col_max
-         M = np.exp(M)
-         #p = bsxfun(@rdivide, M, sum(M));
-         col_sum = np.sum(M, axis=0)
-         divisor = np.tile(col_sum, (M.shape[0], 1))
-         p= np.divide(M, divisor)
-         reg = (W21_tilde@X_21-W1@X_21).T
-         norm_column = []
-         for i in range(nSam2): 
-             norm_column.append(np.linalg.norm(reg[i,:]))
-         norm_column=np.mat(norm_column)    
-         MM = 1.0/(norm_column+epsilon)
-         #D = np.diag(MM.tolist()[0])#%     D = eye(nSam2); Memory Error
-         cost = -1/nSam2 * train_label.flatten() @ np.log(p.flatten()).T + alpha * np.sum(np.square(W21.flatten()))+beta *np.sum(norm_column[0:])
-         W21_tilde_grad = -1/nSam2 * (train_label[0:nCla1,:] - p[0:nCla1,:]) @ X_21.T + 2*alpha*W21_tilde + beta *(AxdiagB(X_21,MM)@(W21_tilde@X_21-W1@X_21).T).T
-         W21_hat_grad =  -1/nSam2 * (train_label[nCla1,:] - p[nCla1,:]) @ X_21.T + 2*alpha*W21_hat
-         
-         W21_tilde = W21_tilde - eta*W21_tilde_grad
-         W21_hat = W21_hat - eta*W21_hat_grad
-        
-         W21=np.vstack(((W21_tilde),W21_hat))
-         
-         if object_value==[]:
-            object_value = np.vstack((np.tile(object_value, (cost.shape[1],1)).T, cost)) 
-         else:
-            object_value=np.vstack(((object_value),cost))
-            
-    softmaxModel = W21
-    return softmaxModel
diff --git a/StreamLearn/tests/tests_SAFC/predict.py b/StreamLearn/tests/tests_SAFC/predict.py
deleted file mode 100644
index 9b75edfdaca514f31a992128e6a05fb13f7bc54b..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/predict.py
+++ /dev/null
@@ -1,49 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-import numpy as np
-import math
-from sklearn.metrics import *
-
-#%%
-# 函数、Predict函数
-def Predict(softmaxModel, test_data,test_label_ori):
-##%% Input:
-#% softmaxModel: the learned multi-class classifier.
-#% test_data: the d*n input data matrix, where each column test_data(:, i) corresponds to
-#%        a single test set
-#% test_label: the c*n input label matrix
-#% our code should produce the prediction matrix 
-##%% Output:
-#% pred: a n-dimension vector, where pred(i) is the prediction for the i-th test_data
-#% acc: the testing accuracy.
-    pred = np.zeros((1, np.shape(test_data)[1]))
-##%% ---------- SoftmaxPredict --------------------------------------
-#%  Compute pred assuming that the labels start from 1.    
-    test_data = test_data[0:np.shape(softmaxModel)[1],:]
-    #M = bsxfun(@minus,,max(softmaxModel*test_data,[],1));
-    softmaxModel_test_data = softmaxModel@test_data
-    col_max = np.max(softmaxModel_test_data, axis=0)
-    M = softmaxModel_test_data - col_max
-    M = np.exp(M)
-    #predall = bsxfun(@rdivide, M, sum(M))
-    col_sum = np.sum(M, axis=0)
-    divisor = np.tile(col_sum, (M.shape[0], 1))
-    predall= np.divide(M, divisor)
-    pred = np.argmax(predall,axis=0)+1
-    probability = np.max(predall,axis=0)
-    test_label = np.argmax(test_label_ori, axis=0) + 1
-    acc = np.mean(test_label.flatten()== pred.flatten())
-    AUC=roc_auc_score(test_label_ori.tolist(),predall.tolist())
-    F1_weight=f1_score(test_label.tolist(),pred.tolist()[0], average='weighted')
-    F1_macro=f1_score(test_label.tolist(),pred.tolist()[0], average='macro')
-    F1_micro=f1_score(test_label.tolist(),pred.tolist()[0], average='micro')
-    #AUC=1
-    #F1_weight=1
-    #F1_macro=1
-    #F1_micro=1
-    return pred,acc,AUC,F1_weight,F1_macro,F1_micro
\ No newline at end of file
diff --git a/StreamLearn/tests/tests_SAFC/request_import.py b/StreamLearn/tests/tests_SAFC/request_import.py
deleted file mode 100644
index 071fdbad0102cf41a4e0b0e38a72938ea8786474..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/request_import.py
+++ /dev/null
@@ -1,33 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-#本实验需要用到的库
-import time
-
-import numpy as np
-import math
-
-from sklearn.preprocessing import StandardScaler
-from sklearn import model_selection
-from sklearn.metrics import *
-
-from compared_method import *
-from predict import *
-from SAFC import *
-from MetricsPred import *
-from validfunc_both import *
-from Read_to_Python import *
-from SAFC_Stream_Funcs import *
-
-from sklearn import svm
-from sklearn.svm import SVC
-from sklearn.svm import LinearSVC
-
-import scipy.io as io
-from scipy.io import loadmat
-import os
-from joblib import dump,load
\ No newline at end of file
diff --git a/StreamLearn/tests/tests_SAFC/validfunc_both.py b/StreamLearn/tests/tests_SAFC/validfunc_both.py
deleted file mode 100644
index 1fe346ed4ae17652ca9c21ac218e81cf4c2a458a..0000000000000000000000000000000000000000
--- a/StreamLearn/tests/tests_SAFC/validfunc_both.py
+++ /dev/null
@@ -1,42 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Sat Sep 14 2024
-
-@author: zhangxinyue
-"""
-
-import numpy as np
-from SAFC import *
-from sklearn import model_selection
-from sklearn.metrics import *
-from predict import *
-
-#%%
-def validfunc(train_data,train_label,alpha_set,eta,method,W1):
-    n = np.shape(train_data)[1]
-    k_fold = 5
-    kf=model_selection.KFold(n_splits=5,shuffle=False,random_state=None)
-    both_statistic = np.zeros((len(alpha_set),len(alpha_set)))
-    for train_index, test_index in kf.split(train_data.T):
-        x_train, x_test = train_data.T[train_index].T, train_data.T[test_index].T
-        y_train, y_test = train_label.T[train_index].T, train_label.T[test_index].T  
-        for i in range(1,len(alpha_set)+1):
-            for j in range(1,len(alpha_set)+1):
-                alpha = alpha_set[i-1]
-                beta = alpha_set[j-1]
-                if method=="SAFC_D":
-                    w_ours1 = SAFC_D(W1,x_train,y_train,alpha,beta,eta)
-                    both_i_j=Predict(w_ours1,x_test, y_test)[1]+Predict(w_ours1,x_test, y_test)[2]
-                    both_statistic[i-1,j-1]= both_statistic[i-1,j-1]+both_i_j
-                elif method=="SAFC_ID":
-                    w_ours2 = SAFC_ID(W1,x_train,y_train,alpha,beta,eta)
-                    both_i_j=Predict(w_ours2,x_test, y_test)[1]+Predict(w_ours2,x_test, y_test)[2]
-                    both_statistic[i-1,j-1]= both_statistic[i-1,j-1]+both_i_j
-                else:
-                    print("Error!")
-                    
-    alpha_position=int(np.argmax(both_statistic)/len(alpha_set))
-    beta_position=np.argmax(both_statistic)%len(alpha_set)
-    alpha_best = alpha_set[alpha_position]
-    beta_best = alpha_set[beta_position]
-    return alpha_best,beta_best
\ No newline at end of file