modify

industry/industry_predict_everyday_100.py

@@ -36,10 +36,10 @@ def predict_today(file, day, model='10_18d', log=True):
 
     for line in lines:
         train_x = np.array([line[:size - 1]])
-        train_x_tmp = train_x[:,:10*8]
-        train_x_a = train_x_tmp.reshape(train_x.shape[0], 10, 8, 1)
+        train_x_tmp = train_x[:,:11*11]
+        train_x_a = train_x_tmp.reshape(train_x.shape[0], 11, 11, 1)
         # train_x_b = train_x_tmp.reshape(train_x.shape[0], 18, 24)
-        train_x_c = train_x[:,10*8:]
+        train_x_c = train_x[:,11*11:]
 
         result = model.predict([train_x_c, train_x_a, ])
         # print(result, line[-1])
@@ -66,6 +66,6 @@ def predict_today(file, day, model='10_18d', log=True):
 
 
 if __name__ == '__main__':
-    predict_today("D:\\data\\quantization\\industry\\stock13_10d_3D_20200417.log", 20200417, model='107_10d_mix_3D_s_seq.h5', log=True)
+    predict_today("D:\\data\\quantization\\industry\\stock15_10d_3D_20200417.log", 20200417, model='111_10d_mix_3D_s_seq.h5', log=True)
     # join_two_day(20200305, 20200305)
     # check_everyday(20200311, 20200312)

industry/industry_train_100.py

@@ -16,13 +16,14 @@ from keras.callbacks import EarlyStopping
 
 early_stopping = EarlyStopping(monitor='accuracy', patience=5, verbose=2)
 
-epochs= 50
+epochs= 33
 # size = 24000 #共68W
-file_path = 'D:\\data\\quantization\\industry\\stock13_10d_3D_train1.log'
-model_path = '107_10d_mix_3D_s_seq.h5'
-file_path1='D:\\data\\quantization\\industry\\stock13_10d_3D_train2.log'
-row = 10
-col = 8
+file_path = 'D:\\data\\quantization\\industry\\stock15_10d_3D_train1.log'
+model_path = '111_10d_mix_3D_s_seq.h5'
+file_path1='D:\\data\\quantization\\industry\\stock15_10d_3D_train2.log'
+file_path2='D:\\data\\quantization\\industry\\stock15_10d_3D_train3.log'
+row = 11
+col = 11
 '''
 1   10-3天    10*9    35,103,34-28
 2   10-3/             34,101,35-28    !
@@ -31,6 +32,11 @@ col = 8
 5   18-3 win=3  18*9  32,100,42-30
 6   18-3 win=6  18*9  37,100,38,27
 7   10-3 win=3 amount涨幅 10*8 37,100,40-27
+8   10-3 ma5 10*9     36,100,40-27     
+9   11-3 win=3 low+high  11*11      31,100,42-24   
+10  11-3 win=4 low+high  11*11      35,100,43-29
+11  11-3 win=6 low+high  11*11      38,100,41     !
+12  12-5 win=6 ma5       12*11      37,100,42-27
 '''
 
 def read_data(path, path1=file_path1):
@@ -45,6 +51,11 @@ def read_data(path, path1=file_path1):
             line = eval(x.strip())
             lines.append(line)
 
+    with open(file_path2) as f:
+        for x in f.readlines(): #680000
+            line = eval(x.strip())
+            lines.append(line)
+
     random.shuffle(lines)
     print('读取数据完毕')
 
@@ -81,7 +92,7 @@ def create_mlp(dim, regress=False):
     model.add(Dense(256, input_dim=dim, activation="relu"))
     model.add(Dropout(0.2))
     model.add(Dense(256, activation="relu"))
-    model.add(Dense(256, activation="relu"))
+    # model.add(Dense(256, activation="relu"))
     model.add(Dense(128, activation="relu"))
 
     # check to see if the regression node should be added
@@ -136,7 +147,7 @@ def create_cnn(width, height, depth, size=48, kernel_size=(5, 6), regress=False,
 # create the MLP and CNN models
 mlp = create_mlp(train_x_c.shape[1], regress=False)
 # cnn_0 = create_cnn(18, 20, 1, kernel_size=(3, 3), size=90, regress=False, output=96)       # 31 97 46
-cnn_0 = create_cnn(row, col, 1, kernel_size=(3, col), size=96, regress=False, output=96)         # 29 98 47
+cnn_0 = create_cnn(row, col, 1, kernel_size=(6, col), size=88, regress=False, output=88)         # 29 98 47
 # cnn_0 = create_cnn(18, 20, 1, kernel_size=(9, 9), size=90, regress=False, output=96)         # 28 97 53
 # cnn_0 = create_cnn(18, 20, 1, kernel_size=(3, 20), size=90, regress=False, output=96)
 # cnn_1 = create_cnn(18, 20, 1, kernel_size=(18, 10), size=80, regress=False, output=96)

mix/mix_predict_500.py

@@ -22,15 +22,16 @@ def _score(fact, line):
     up_error = 0
 
     if fact[0] == 1:
-        up_right = up_right + 1.12
+        up_right = up_right + 1.1
     elif fact[1] == 1:
-        up_right = up_right + 1.04
+        up_error = up_error + 0.4
+        up_right = up_right + 1.03
     elif fact[2] == 1:
-        up_error = up_error + 0.3
-        up_right = up_right + 0.98
+        up_error = up_error + 0.7
+        up_right = up_right + 0.96
     else:
         up_error = up_error + 1
-        up_right = up_right + 0.94
+        up_right = up_right + 0.9
     return up_right,up_error
 
 
@@ -63,15 +64,15 @@ def predict(file_path='', model_path='15min_dnn_seq.h5', idx=-1, row=18, col=20)
             if r[0] > 0.5 or r[1] > 0.5:
                 pass
         else:
-            if r[0] > 0.6 :
+            if r[0] > 0.5 :
                 tmp_right,tmp_error = _score(fact, lines[i])
                 up_right = tmp_right + up_right
                 up_error = tmp_error + up_error
                 up_num = up_num + 1
-            elif r[2] > 0.6 or r[3] > 0.6:
+            elif r[2] > 0.5 or r[3] > 0.5:
                 if fact[0] == 1:
                     down_error = down_error + 1
-                    down_right = down_right + 1.14
+                    down_right = down_right + 1.1
                 elif fact[1] == 1:
                     down_error = down_error + 0.5
                     down_right = down_right + 1.04
@@ -94,7 +95,8 @@ if __name__ == '__main__':
     # predict(file_path='D:\\data\\quantization\\stock181_18d_test.log', model_path='181_18d_mix_6D_ma5_s_seq.h5')
     # predict(file_path='D:\\data\\quantization\\stock217_18d_train1.log', model_path='218_18d_mix_5D_ma5_s_seq.h5', row=18, col=18)
     # predict(file_path='D:\\data\\quantization\\stock400_18d_train1.log', model_path='400_18d_mix_5D_ma5_s_seq.h5', row=18, col=18)
-    predict(file_path='D:\\data\\quantization\\stock507_28d_train1.log', model_path='507_28d_mix_5D_ma5_s_seq.h5', row=28, col=19)
+    predict(file_path='D:\\data\\quantization\\stock517_28d_train1.log', model_path='517_28d_mix_3D_ma5_s_seq.h5', row=28, col=16)
     # predict(file_path='D:\\data\\quantization\\stock6_test.log', model_path='15m_dnn_seq.h5')
     # multi_predict(model='15_18d')
-    # predict_today(20200229, model='11_18d')
+    # predict_today(20200229, model='11_18d')
+    # predict(file_path='D:\\data\\quantization\\stock513_28d_train1.log', model_path='513_28d_mix_3D_ma5_s_seq.h5', row=28, col=16)

mix/mix_predict_518.py

@@ -2,6 +2,7 @@
 import numpy as np
 from keras.models import load_model
 import joblib
+import random
 
 
 def read_data(path):
@@ -17,31 +18,82 @@ def read_data(path):
     return np.array(train_x, dtype=np.float32),np.array(train_y, dtype=np.float32),lines
 
 
-def _score(fact, line):
+def _score(fact,):
     up_right = 0
     up_error = 0
 
     if fact[0] == 1:
         up_right = up_right + 1.1
     elif fact[1] == 1:
-        up_error = up_error + 0.2
-        up_right = up_right + 1.04
+        up_error = up_error + 0.4
+        up_right = up_right + 1.03
     elif fact[2] == 1:
-        up_error = up_error + 0.5
-        up_right = up_right + 0.98
+        up_error = up_error + 0.7
+        up_right = up_right + 0.96
     else:
         up_error = up_error + 1
-        up_right = up_right + 0.94
+        up_right = up_right + 0.9
     return up_right,up_error
 
 
+def random_predict(file_path=''):
+    test_x,test_y,lines=read_data(file_path)
+    a=0
+    b=0
+    c=0
+    d=0
+    R=0
+
+    up_num = 0
+    up_error = 0
+    up_right = 0
+    down_num = 0
+    down_error = 0
+    down_right = 0
+
+    for x in test_y:
+        xx = random.uniform(0, 4)
+        if xx > 3:
+            if x[0] == 1:
+                R = R + 1
+            tmp_right,tmp_error = _score(x,)
+            up_num = up_num + 1
+            up_right = up_right + tmp_right
+            up_error = tmp_error + up_error
+        elif xx > 2:
+            if x[1] == 1:
+                R = R + 1
+        elif xx > 1:
+            if x[2] == 1:
+                R = R + 1
+        elif xx > 0:
+            if x[3] == 1:
+                R = R + 1
+
+        if x[0] == 1:
+            a = a + 1
+        elif x[1] == 1:
+            b = b + 1
+        elif x[2] == 1:
+            c = c + 1
+        else:
+            d = d + 1
+
+    if up_num == 0:
+        up_num = 1
+    if down_num == 0:
+        down_num = 1
+
+    print(R/(a + b + c +d), up_right/up_num, up_error/up_num)
+
+
 def predict(file_path='', model_path='15min_dnn_seq.h5', idx=-1, row=18, col=20):
     test_x,test_y,lines=read_data(file_path)
 
     test_x_a = test_x[:,:row*col]
     test_x_a = test_x_a.reshape(test_x.shape[0], row, col, 1)
-    # test_x_b = test_x[:, row*col:row*col+18*2]
-    # test_x_b = test_x_b.reshape(test_x.shape[0], 18, 2, 1)
+    # test_x_b = test_x[:, 18*col:row*col]
+    # test_x_b = test_x_b.reshape(test_x.shape[0], 10, col, 1)
     test_x_c = test_x[:,row*col:]
 
     model=load_model(model_path)
@@ -56,16 +108,19 @@ def predict(file_path='', model_path='15min_dnn_seq.h5', idx=-1, row=18, col=20)
     down_right = 0
     i = 0
     result = model.predict([test_x_c, test_x_a, ])
+
+
     win_dnn = []
     for r in result:
         fact = test_y[i]
+        xx = random.uniform(0, 4)
 
         if idx in [-2]:
             if r[0] > 0.5 or r[1] > 0.5:
                 pass
         else:
-            if r[0] > 0.5 :
-                tmp_right,tmp_error = _score(fact, lines[i])
+            if r[1] > 0.6 :
+                tmp_right,tmp_error = _score(fact,)
                 up_right = tmp_right + up_right
                 up_error = tmp_error + up_error
                 up_num = up_num + 1
@@ -74,12 +129,13 @@ def predict(file_path='', model_path='15min_dnn_seq.h5', idx=-1, row=18, col=20)
                     down_error = down_error + 1
                     down_right = down_right + 1.1
                 elif fact[1] == 1:
-                    down_error = down_error + 0.5
+                    down_error = down_error + 0.7
                     down_right = down_right + 1.04
                 elif fact[2] == 1:
-                    down_right = down_right + 0.98
+                    down_error = down_error + 0.3
+                    down_right = down_right + 0.94
                 else:
-                    down_right = down_right + 0.92
+                    down_right = down_right + 0.9
                 down_num = down_num + 1
 
         i = i + 1
@@ -92,11 +148,8 @@ def predict(file_path='', model_path='15min_dnn_seq.h5', idx=-1, row=18, col=20)
 
 
 if __name__ == '__main__':
-    # predict(file_path='D:\\data\\quantization\\stock181_18d_test.log', model_path='181_18d_mix_6D_ma5_s_seq.h5')
-    # predict(file_path='D:\\data\\quantization\\stock217_18d_train1.log', model_path='218_18d_mix_5D_ma5_s_seq.h5', row=18, col=18)
-    # predict(file_path='D:\\data\\quantization\\stock400_18d_train1.log', model_path='400_18d_mix_5D_ma5_s_seq.h5', row=18, col=18)
-    predict(file_path='D:\\data\\quantization\\stock517_28d_train1.log', model_path='517_28d_mix_3D_ma5_s_seq.h5', row=28, col=16)
-    # predict(file_path='D:\\data\\quantization\\stock6_test.log', model_path='15m_dnn_seq.h5')
-    # multi_predict(model='15_18d')
-    # predict_today(20200229, model='11_18d')
-    # predict(file_path='D:\\data\\quantization\\stock513_28d_train1.log', model_path='513_28d_mix_3D_ma5_s_seq.h5', row=28, col=16)
+    xx = random.uniform(0,4)
+    # predict(file_path='D:\\data\\quantization\\stock513_28d_train1.log', model_path='513_28d_mix_3D_ma5_s_seq.h5', row=28, col=16)
+    predict(file_path='D:\\data\\quantization\\stock538_28d_train1.log', model_path='539_28d_mix_5D_ma5_s_seq.h5', row=28, col=17)
+    # random_predict(file_path='D:\\data\\quantization\\week101_18d_test.log')
+    # random_predict(file_path='D:\\data\\quantization\\stock537_28d_train1.log')

mix/mix_predict_by_day_324.py

@@ -60,9 +60,9 @@ def predict(file_path='', model_path='15min_dnn_seq', rows=18, cols=18):
             if result[0][0]> 0.5:
                 up_num = up_num + ratio
             elif result[0][1] > 0.5:
-                up_num = up_num + 0.01*ratio
+                up_num = up_num + 0.4*ratio
             elif result[0][2] > 0.5:
-                down_num = down_num + 0.01*ratio
+                down_num = down_num + 0.4*ratio
             else:
                 down_num = down_num + ratio
 
@@ -77,10 +77,23 @@ def predict(file_path='', model_path='15min_dnn_seq', rows=18, cols=18):
                 x3 = x3 + 1
 
         # print(key, int(up_num), int(down_num), (down_num*1.2 + 2)/(up_num*1.2 + 2), )
-        print(key, x0, x1, x2,x3, (down_num*1.2 + 2)/(up_num*1.2 + 2))
+        print(key, x0, x1, x2,x3, (down_num*1.5 + 2)/(up_num*1.2 + 2))
 
 
 if __name__ == '__main__':
-    # predict(file_path='D:\\data\\quantization\\stock324_28d_3D_20200415.log', model_path='324_28d_mix_5D_ma5_s_seq', rows=28, cols=18)
+    # predict(file_path='D:\\data\\quantization\\stock6_5_test.log', model_path='5d_dnn_seq.h5')
+    # predict(file_path='D:\\data\\quantization\\stock9_18_20200220.log', model_path='18d_dnn_seq.h5')
+    # predict(file_path='D:\\data\\quantization\\stock9_18_2.log', model_path='18d_dnn_seq.h5')
+    # predict(file_path='D:\\data\\quantization\\stock16_18d_20200310.log', model_path='16_18d_mix_seq')
+    # predict(file_path='D:\\data\\quantization\\stock196_18d_20200326.log', model_path='196_18d_mix_6D_ma5_s_seq')
 
-    predict(file_path='D:\\data\\quantization\\stock324_28d_3D_20191221.log', model_path='324_28d_mix_5D_ma5_s_seq', rows=28, cols=18)
+
+    # predict(file_path='D:\\data\\quantization\\stock321_28d_5D_20200429.log', model_path='321_28d_mix_5D_ma5_s_seq_2', rows=28, cols=20)
+
+    predict(file_path='D:\\data\\quantization\\stock327_28d_20200429.log', model_path='327_28d_mix_5D_ma5_s_seq', rows=28, cols=20)
+
+
+
+    # predict(file_path='D:\\data\\quantization\\stock9_18_4.log', model_path='18d_dnn_seq.h5')
+    # predict(file_path='D:\\data\\quantization\\stock324_28d_3D_20200414_A.log', model_path='324_28d_mix_5D_ma5_s_seq', rows=28, cols=18)
+    # predict(file_path='D:\\data\\quantization\\stock324_28d_3D_20200414_A.log', model_path='603_30d_mix_5D_ma5_s_seq', rows=30, cols=19)

mix/mix_predict_everyday_500.py

@@ -36,16 +36,16 @@ def predict_today(file, day, model='10_18d', log=True):
 
     for line in lines:
         train_x = np.array([line[:size - 1]])
-        train_x_tmp = train_x[:,:28*20]
-        train_x_a = train_x_tmp.reshape(train_x.shape[0], 28, 20, 1)
+        train_x_tmp = train_x[:,:28*16]
+        train_x_a = train_x_tmp.reshape(train_x.shape[0], 28, 16, 1)
         # train_x_b = train_x_tmp.reshape(train_x.shape[0], 18, 24)
-        train_x_c = train_x[:,28*20:]
+        train_x_c = train_x[:,28*16:]
 
         result = model.predict([train_x_c, train_x_a, ])
         # print(result, line[-1])
         stock = code_table.find_one({'ts_code':line[-1][0]})
 
-        if result[0][0] > 0.5 and stock['sw_industry'] in industry_list:
+        if result[0][0] > 0.5:
             if line[-1][0].startswith('688'):
                 continue
             # 去掉ST
@@ -67,12 +67,17 @@ def predict_today(file, day, model='10_18d', log=True):
             print(stock['ts_code'], stock['name'], '买入')
             O_list.append([stock['ts_code'], stock['name']])
 
-            if log is True:
-                with open('D:\\data\\quantization\\predict\\' + str(day) + '_mix500.txt', mode='a', encoding="utf-8") as f:
+            if result[0][0] > 0.9:
+                R_list.append([stock['ts_code'], stock['name']])
+
+                with open('D:\\data\\quantization\\predict\\' + str(day) + '_mix537.txt', mode='a', encoding="utf-8") as f:
                     f.write(str(line[-1]) + ' ' + stock['name'] + ' ' + stock['sw_industry'] + ' ' + str(concept_detail_list) + ' ' + str(result[0][0]) + '\n')
 
-        elif result[0][1] > 0.5 or result[0][2] > 0.5 :
+        elif result[0][1] > 0.5:
             pass
+        elif result[0][2] > 0.5:
+            if stock['ts_code'] in holder_stock_list:
+                print(stock['ts_code'], stock['name'], '警告危险')
         elif result[0][3] > 0.5:
             if stock['ts_code'] in holder_stock_list or stock['ts_code'] in zixuan_stock_list:
                 print(stock['ts_code'], stock['name'], '赶紧卖出')
@@ -81,8 +86,12 @@ def predict_today(file, day, model='10_18d', log=True):
 
     # print(gainian_map)
     # print(hangye_map)
-    random.shuffle(O_list)
-    print(O_list[:3])
+    # random.shuffle(O_list)
+    # print(O_list[:3])
+
+    random.shuffle(R_list)
+    print('----ROE----')
+    print(R_list[:])
 
 
 if __name__ == '__main__':
@@ -90,6 +99,9 @@ if __name__ == '__main__':
     # predict(file_path='D:\\data\\quantization\\stock6_test.log', model_path='15m_dnn_seq.h5')
     # multi_predict()
     # 策略B
-    predict_today("D:\\data\\quantization\\stock505_28d_20200415.log", 20200415, model='505_28d_mix_5D_ma5_s_seq.h5', log=True)
+    # predict_today("D:\\data\\quantization\\stock505_28d_20200416.log", 20200416, model='505_28d_mix_5D_ma5_s_seq.h5', log=True)
+    predict_today("D:\\data\\quantization\\stock517_28d_20200429.log", 20200429, model='517_28d_mix_3D_ma5_s_seq.h5', log=True)
+    # predict_today("D:\\data\\quantization\\stock538_28d_20200205.log", 20200205, model='539_28d_mix_5D_ma5_s_seq.h5', log=True)
+
     # join_two_day(20200305, 20200305)
     # check_everyday(20200311, 20200312)

mix/mix_train_300.py

@@ -18,14 +18,14 @@ from keras.callbacks import EarlyStopping
 
 early_stopping = EarlyStopping(monitor='accuracy', patience=5, verbose=2)
 
-epochs= 77
-size = 440000 #共68W
-file_path = 'D:\\data\\quantization\\stock324_28d_train2.log'
-model_path = '324_28d_mix_5D_ma5_s_seq.h5'
-file_path1='D:\\data\\quantization\\stock324_28d_test.log'
-file_path2='D:\\data\\quantization\\stock324_28d_train1.log'
+epochs= 60
+size = 444000 #共68W
+file_path = 'D:\\data\\quantization\\stock327_28d_train2.log'
+model_path = '327_28d_mix_5D_ma5_s_seq.h5'
+file_path1='D:\\data\\quantization\\stock327_28d_test.log'
+file_path2='D:\\data\\quantization\\stock327_28d_train1.log'
 row = 28
-col = 18
+col = 20
 '''
 30d+ma5+流通市值>40
 0 ROC     30*18           38,100,17
@@ -39,10 +39,15 @@ col = 18
 28d+ma5+5+流通市值>10
 21 DMI     28*20          43,102,9  非常好    46,102,8
 22 MACD    28*19          46,102,9
+25 DMI ma5+涨幅int 28*18  40,102,8/14
+26 DMI 向量化   28*22     41,101,14,14
+27 DMI 向量化 修正   28*20     41,101,14,14
+
 1d close
 23 DMI     28*20          34,97,36
 3d close 去掉ma的两个字段
 24 DMI     28*18          41,96,42-13
+
        
 30d+close
 4 ROC     30*18           
@@ -62,7 +67,7 @@ def read_data(path, path1=file_path1):
             lines.append(line)
 
     with open(path1) as f:
-        for x in range(50000): #6w
+        for x in range(30000): #6w
             line = eval(f.readline().strip())
             lines.append(line)
 

mix/mix_train_500.py

@@ -18,22 +18,33 @@ from keras.callbacks import EarlyStopping
 
 early_stopping = EarlyStopping(monitor='accuracy', patience=5, verbose=2)
 
-epochs= 44
-size = 440000 #共68W
-file_path = 'D:\\data\\quantization\\stock507_28d_train2.log'
-model_path = '507_28d_mix_5D_ma5_s_seq.h5'
-file_path1='D:\\data\\quantization\\stock507_28d_test.log'
+epochs= 30
+size = 470000 #共68W
+file_path = 'D:\\data\\quantization\\stock517_28d_train2.log'
+model_path = '517_28d_mix_3D_ma5_s_seq.h5'
+file_path1='D:\\data\\quantization\\stock517_28d_test.log'
 row = 28
-col = 19
+col = 16
 '''
 0    dmi            28*20      38,98,51/5     下跌预判非常准                                    
 1    macd           28*19      41,98,53/8  
 2    dmi-对大盘对比 28*20      35,99,46/17 
 3    5d-dmi-对大盘对比 28*20   42,99,39/10
 4    3d-dmi-对大盘对比 28*20   40,99,39/07   
-5    3d-beta1                  55,99,39/07    ==> 用这个
+5    3d-beta1                  55,99,52/07    当前用这个 
 6    3d-ma20                   40,99,41/07
 7    3d-macd   28*19           55,99,40/07
+8    3d-市值>30  28*20         57,101,36/14   最高价  用这个！
+9    3d-市值>30  28*20         57,99,31/08   收盘最高价 
+10   5d-市值>30  28*20   收盘最高价  
+11   5d-市值>30  28*20   ma5
+12   5d-极简     28*16 有ma5,ma20  46,102,16/26
+13   3d-最高价   28*16         57,101,39,16
+14   5d-极简-最高价     28*16   40,102,30-34
+15   5d+dmi+最高价      28*20   40,102,31-34    
+16   同12,14,参数11,10  28*16   38,102,29-36
+17   同上参数11,6       28*16   39,101,46-35  !    33,100,48
+
 '''
 
 def read_data(path, path1=file_path1):
@@ -44,7 +55,7 @@ def read_data(path, path1=file_path1):
             lines.append(line)
 
     with open(path1) as f:
-        for x in range(33000): #6w
+        for x in range(30000): #6w
             line = eval(f.readline().strip())
             lines.append(line)
 

mix/mix_train_600.py

@@ -18,15 +18,16 @@ early_stopping = EarlyStopping(monitor='accuracy', patience=5, verbose=2)
 
 epochs= 44
 size = 440000 #共68W
-file_path = 'D:\\data\\quantization\\stock603_30d_train2.log'
-model_path = '603_30d_mix_5D_ma5_s_seq.h5'
-file_path1='D:\\data\\quantization\\stock603_30d_test.log'
+file_path = 'D:\\data\\quantization\\stock604_30d_train2.log'
+model_path = '604_30d_mix_5D_ma5_s_seq.h5'
+file_path1='D:\\data\\quantization\\stock604_30d_test.log'
 row = 30
 col = 19
 '''
 1   macd+beta1 盈利       30*19             64,99,31-31
 2   macd+beta1 亏损       30*19
-3   macd+减少指数参数     30*19             train1:69,99,32-31 |train3:62,100,23-39
+3   macd+减少指数参数 最高价     30*19             train1:69,99,32-31 |train3:62,100,23-39 !!    >>   39,97,54
+4   macd+减少指数参数 收盘最高价     30*19         53,98,38-23    X
 '''
 
 def read_data(path, path1=file_path1):

mix/stock_source.py

@@ -5,7 +5,8 @@ mysql_handler = Mysql()
 
 zixuan_stock_list = [
     # 医疗
-    '603990.SH', '300759.SZ', '300347.SZ','002421.SZ','300168.SZ','002432.SZ','300074.SZ','300677.SZ',
+    '603990.SH', '300759.SZ', '300347.SZ','002421.SZ','300168.SZ','002432.SZ','300074.SZ','300677.SZ','002223.SZ', '300253.SZ', '300629.SZ',
+    '002603.SZ', '300147.SZ', '002107.SZ',
     # 5G
     '300003.SZ', '600498.SH', '300310.SZ', '603912.SH', '603220.SH', '300602.SZ', '600260.SH', '002463.SZ','300738.SZ','002402.SZ',
     # 车联网
@@ -15,12 +16,18 @@ zixuan_stock_list = [
     # 特高压
     '300341.SZ', '300670.SZ', '300018.SZ', '600268.SH', '002879.SZ','002028.SZ','300477.SZ',
     # 基础建设
-    '603568.SH', '000967.SZ', '603018.SH','002062.SZ',
+    '603568.SH', '000967.SZ', '603018.SH','002062.SZ','600452.SH',
     # 华为
     '300687.SZ','002316.SZ','300339.SZ','300378.SZ','300020.SZ','300634.SZ','002570.SZ', '300766.SZ',
 
-    '002555.SZ','600585.SH','600276.SH','002415.SZ','000651.SZ',
+]
 
+zeng_stock_list = [
+    # 业绩
+    '300097.SZ', '000625.SZ', '000876.SZ', '603258.SH', '002973.SZ', '300390.SZ', '002030.SZ', '002041.SZ', '000767.SZ', '002214.SZ', '002869.SZ',
+    '300541.SZ', '002185.SZ',
+    '002555.SZ','600585.SH','600276.SH','002415.SZ','000651.SZ',
+    '601800.SH', '601108.SH', '601186.SH',
 ]
 
 ROE_stock_list =   [                      # ROE
@@ -42,11 +49,12 @@ ROE_stock_list =   [                      # ROE
 
 
 holder_stock_list = [
-    '600498.SH', '002223.SZ',
-    '600496.SH', '300682.SZ','601162.SH','002401.SZ','601111.SH',
-    '000851.SZ','300639.SZ','603990.SH','603003.SH','603628.SH','601186.SH',
-    '600196.SH', '300003.SZ','300748.SZ','603638.SZ',
-    '601211.SH'
+    '002415.SZ','002214.SZ', '002459.SZ'
+
+    '300639.SZ','300003.SZ', '300185.SZ',
+
+    '600196.SH', '601108.SH', '600057.SH',
+    '601211.SH', '600452.SH', '603638.SH',
 ]
 
 

stock/kmeans.py

@@ -8,7 +8,7 @@ import joblib
 def read_data(path):
     lines = []
     with open(path) as f:
-        for x in range(600000):
+        for x in range(400000):
             line = eval(f.readline().strip())
             # if line[-1][0] == 1 or line[-1][1] == 1:
             lines.append(line)
@@ -84,7 +84,68 @@ def class_fic_k(file_path=''):
     print(estimator.predict(v_x[10:20]))
 
 
+def class_fic_roc(file_path=''):
+    lines = read_data(file_path)
+    print('读取数据完毕')
+    size = len(lines[0])
+    x_list = []
+
+    length = 18  # 周期是多少
+    j = 10
+
+    for s in lines:
+        tmp_list = []
+        for x in range(0, length):
+            tmp_list = tmp_list + s[x*j+8:x*j+10]
+        x_list.append(tmp_list)
+    train_x = np.array(x_list)
+
+    v_x = train_x.reshape(train_x.shape[0], 2*length)
+
+    estimator = KMeans(n_clusters=8, random_state=73011)
+    estimator.fit(v_x)
+    label_pred = estimator.labels_  # 获取聚类标签
+    centroids = estimator.cluster_centers_
+    joblib.dump(estimator , 'km_k_roc_8.pkl')
+
+    print(estimator.predict(v_x[:10]))
+
+    estimator = joblib.load('km_k_roc_8.pkl')
+    print(estimator.predict(v_x[10:20]))
+
+
+def class_fic_dapan(file_path=''):
+    lines = read_data(file_path)
+    print('读取数据完毕')
+    size = len(lines[0])
+    x_list = []
+
+    length = 18  # 周期是多少
+    j = 24
+
+    for s in lines:
+        tmp_list = []
+        for x in range(0, length):
+            tmp_list = tmp_list + s[x*j+16:x*j+17]
+        x_list.append(tmp_list)
+    train_x = np.array(x_list)
+
+    v_x = train_x.reshape(train_x.shape[0], length)
+
+    estimator = KMeans(n_clusters=8, random_state=82430)
+    estimator.fit(v_x)
+    label_pred = estimator.labels_  # 获取聚类标签
+    centroids = estimator.cluster_centers_
+    joblib.dump(estimator , 'km_k_dapan_8.pkl')
+
+    print(estimator.predict(v_x[:10]))
+
+    estimator = joblib.load('km_k_dapan_8.pkl')
+    print(estimator.predict(v_x[10:20]))
+
+
 if __name__ == '__main__':
     # class_fic(file_path="D:\\data\\quantization\\stock2_10.log")
     # class_fic_k(file_path="D:\\data\\quantization\\stock10_18_train.log")
-    class_fic_dmi(file_path="D:\\data\\quantization\\stock12_18d_train.log")
+    # class_fic_dmi(file_path="D:\\data\\quantization\\stock12_18d_train.log")
+    class_fic_dapan(file_path="D:\\data\\quantization\\stock160_18d_train.log")