mix(cnn+dnn)训练 16数据最好

mix/lstm_predict.py

@@ -45,14 +45,14 @@ def predict(file_path='', model_path='15min_dnn_seq.h5', idx=-1):
     test_x,test_y,lines=read_data(file_path)
     print('Load data success')
 
-    test_x_a = test_x[:,:18*16]
-    test_x_a = test_x_a.reshape(test_x.shape[0], 18, 16)
-    test_x_b = test_x[:, 18*16:18*16+10*18]
-    test_x_b = test_x_b.reshape(test_x.shape[0], 18, 10, 1)
-    test_x_c = test_x[:,18*16+10*18:]
+    test_x_a = test_x[:,:18*24]
+    test_x_a = test_x_a.reshape(test_x.shape[0], 18, 24)
+    # test_x_b = test_x[:, 18*16:18*16+10*18]
+    # test_x_b = test_x_b.reshape(test_x.shape[0], 18, 10, 1)
+    test_x_c = test_x[:,18*24:]
 
     model=load_model(model_path)
-    score = model.evaluate([test_x_c, test_x_a, test_x_b], test_y)
+    score = model.evaluate([test_x_c, test_x_a, ], test_y)
     print('MIX', score)
 
     up_num = 0
@@ -62,7 +62,7 @@ def predict(file_path='', model_path='15min_dnn_seq.h5', idx=-1):
     down_error = 0
     down_right = 0
     i = 0
-    result=model.predict([test_x_c, test_x_a, test_x_b])
+    result=model.predict([test_x_c, test_x_a, ])
     win_dnn = []
     for r in result:
         fact = test_y[i]
@@ -102,7 +102,7 @@ def predict(file_path='', model_path='15min_dnn_seq.h5', idx=-1):
 
 
 if __name__ == '__main__':
-    predict(file_path='D:\\data\\quantization\\stock19_18d_test.log', model_path='19_18d_lstm_seq.h5')
+    predict(file_path='D:\\data\\quantization\\stock17_18d_test.log', model_path='17_18d_lstm_seq.h5')
     # 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')

mix/lstm_train.py

@@ -20,9 +20,8 @@ from keras.callbacks import EarlyStopping
 
 early_stopping = EarlyStopping(monitor='accuracy', patience=5, verbose=2)
 
-epochs= 40
-size = 80000
-
+epochs= 200
+size = 380000
 
 def read_data(path):
     lines = []
@@ -50,13 +49,13 @@ def read_data(path):
     return X_resampled,y_resampled,np.array(test_x),np.array(test_y)
 
 
-train_x,train_y,test_x,test_y=read_data("D:\\data\\quantization\\stock19_18d_train_1.log")
+train_x,train_y,test_x,test_y=read_data("D:\\data\\quantization\\stock17_18d_train.log")
 
-train_x_a = train_x[:,:18*16]
-train_x_a = train_x_a.reshape(train_x.shape[0], 18, 16)
-train_x_b = train_x[:, 18*16:18*16+10*18]
-train_x_b = train_x_b.reshape(train_x.shape[0], 18, 10, 1)
-train_x_c = train_x[:,18*16+10*18:]
+train_x_a = train_x[:,:18*24]
+train_x_a = train_x_a.reshape(train_x.shape[0], 18, 24)
+# train_x_b = train_x[:, 18*24:18*16+10*18]
+# train_x_b = train_x_b.reshape(train_x.shape[0], 18, 10, 1)
+train_x_c = train_x[:,18*24:]
 
 
 def create_mlp(dim, regress=False):
@@ -127,7 +126,7 @@ def create_lstm(sample, timesteps, input_dim):
 
     x = inputs
 
-    x = LSTM(units = 32, input_shape=(18, 16), dropout=0.2
+    x = LSTM(units = 64, input_shape=(timesteps, input_dim), dropout=0.2
                )(x)
     # x = LSTM(16*16, return_sequences=False)
     # x = Activation("relu")(x)
@@ -143,26 +142,26 @@ def create_lstm(sample, timesteps, input_dim):
 
 # create the MLP and CNN models
 mlp = create_mlp(train_x_c.shape[1], regress=False)
-cnn_0 = create_lstm(train_x_a.shape[1], 18, 16)
-cnn_1 = create_cnn(18, 10, 1, kernel_size=(3, 5), filters=32, regress=False, output=120)
+cnn_0 = create_lstm(train_x_a.shape[1], 18, 24)
+# cnn_1 = create_cnn(18, 10, 1, kernel_size=(3, 5), filters=32, regress=False, output=120)
 
 # create the input to our final set of layers as the *output* of both
 # the MLP and CNN
-combinedInput = concatenate([mlp.output, cnn_0.output, cnn_1.output])
+combinedInput = concatenate([mlp.output, cnn_0.output,])
 
 # our final FC layer head will have two dense layers, the final one
 # being our regression head
-x = Dense(888, activation="relu", kernel_regularizer=regularizers.l1(0.003))(combinedInput)
+x = Dense(666, activation="relu", kernel_regularizer=regularizers.l1(0.003))(combinedInput)
 x = Dropout(0.2)(x)
-x = Dense(888, activation="relu")(x)
-x = Dense(888, activation="relu")(x)
+x = Dense(666, activation="relu")(x)
+x = Dense(666, activation="relu")(x)
 # 在建设一层
 x = Dense(5, activation="softmax")(x)
 
 # our final model will accept categorical/numerical data on the MLP
 # input and images on the CNN input, outputting a single value (the
 # predicted price of the house)
-model = Model(inputs=[mlp.input, cnn_0.input, cnn_1.input], outputs=x)
+model = Model(inputs=[mlp.input, cnn_0.input,], outputs=x)
 
 
 print("Starting training ")
@@ -178,7 +177,7 @@ model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=['accuracy
 # train the model
 print("[INFO] training model...")
 model.fit(
-    [train_x_c, train_x_a, train_x_b], train_y,
+    [train_x_c, train_x_a, ], train_y,
     # validation_data=([testAttrX, testImagesX], testY),
     # epochs=int(3*train_x_a.shape[0]/1300),
     epochs=epochs,
@@ -186,20 +185,20 @@ model.fit(
     callbacks=[early_stopping]
 )
 
-test_x_a = test_x[:,:18*16]
-test_x_a = test_x_a.reshape(test_x.shape[0], 18, 16)
-test_x_b = test_x[:, 18*16:18*16+10*18]
-test_x_b = test_x_b.reshape(test_x.shape[0], 18, 10, 1)
-test_x_c = test_x[:,18*16+10*18:]
+test_x_a = test_x[:,:18*24]
+test_x_a = test_x_a.reshape(test_x.shape[0], 18, 24)
+# test_x_b = test_x[:, 18*16:18*16+10*18]
+# test_x_b = test_x_b.reshape(test_x.shape[0], 18, 10, 1)
+test_x_c = test_x[:,18*24:]
 
 # make predictions on the testing data
 print("[INFO] predicting house prices...")
-score  = model.evaluate([test_x_c, test_x_a, test_x_b], test_y)
+score  = model.evaluate([test_x_c, test_x_a], test_y)
 
 print(score)
 print('Test score:', score[0])
 print('Test accuracy:', score[1])
 
-path="19_18d_lstm_seq.h5"
+path="17_18d_lstm_seq.h5"
 model.save(path)
 model=None

mix/mix_predict_1.py

@@ -0,0 +1,106 @@
+# -*- encoding:utf-8 -*-
+import numpy as np
+from keras.models import load_model
+import joblib
+
+
+def read_data(path):
+    lines = []
+    with open(path) as f:
+        for line in f.readlines()[:]:
+            line = eval(line.strip())
+            if line[-2][0].startswith('0') or line[-2][0].startswith('3'):
+                lines.append(line)
+
+    size = len(lines[0])
+    train_x=[s[:size - 2] for s in lines]
+    train_y=[s[size-1] for s in lines]
+    return np.array(train_x),np.array(train_y),lines
+
+
+def _score(fact, line):
+    with open('dnn_predict_dmi_18d.txt', 'a') as f:
+        f.write(str([line[-2], line[-1]]) + "\n")
+
+    up_right = 0
+    up_error = 0
+
+    if fact[0] == 1:
+        up_right = up_right + 1.12
+    elif fact[1] == 1:
+        up_right = up_right + 1.06
+    elif fact[2] == 1:
+        up_right = up_right + 1
+        up_error = up_error + 0.5
+    elif fact[3] == 1:
+        up_right = up_right + 0.94
+        up_error = up_error + 1
+    else:
+        up_error = up_error + 1
+        up_right = up_right + 0.88
+    return up_right,up_error
+
+
+def predict(file_path='', model_path='15min_dnn_seq.h5', idx=-1):
+    test_x,test_y,lines=read_data(file_path)
+
+    test_x_a = test_x[:,:18*24]
+    test_x_a = test_x_a.reshape(test_x.shape[0], 18, 24, 1)
+    # test_x_b = test_x[:, 18*18:18*18+2*18]
+    # test_x_b = test_x_b.reshape(test_x.shape[0], 18, 2, 1)
+    test_x_c = test_x[:,18*24:]
+
+    model=load_model(model_path)
+    score = model.evaluate([test_x_c, test_x_a], test_y)
+    print('MIX', score)
+
+    up_num = 0
+    up_error = 0
+    up_right = 0
+    down_num = 0
+    down_error = 0
+    down_right = 0
+    i = 0
+    result=model.predict([test_x_c, test_x_a,])
+    win_dnn = []
+    for r in result:
+        fact = test_y[i]
+
+        if idx in [-2]:
+            if r[0] > 0.5 or r[1] > 0.5:
+                pass
+        else:
+            if r[0] > 0.6 or r[1] > 0.6:
+                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[3] > 0.6 or r[4] > 0.6:
+                if fact[0] == 1:
+                    down_error = down_error + 1
+                    down_right = down_right + 1.12
+                elif fact[1] == 1:
+                    down_error = down_error + 1
+                    down_right = down_right + 1.06
+                elif fact[2] == 1:
+                    down_right = down_right + 1
+                elif fact[3] == 1:
+                    down_right = down_right + 0.94
+                else:
+                    down_right = down_right + 0.88
+                down_num = down_num + 1
+
+        i = i + 1
+    if up_num == 0:
+        up_num = 1
+    if down_num == 0:
+        down_num = 1
+    print('MIX', up_right, up_num, up_right/up_num, up_error/up_num, down_right/down_num, down_error/down_num)
+    return win_dnn,up_right/up_num,down_right/down_num
+
+
+if __name__ == '__main__':
+    predict(file_path='D:\\data\\quantization\\stock16_18d_test_close.log', model_path='16_18d_mix_seq.h5')
+    # 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')

mix/mix_predict_by_day.py

@@ -22,10 +22,7 @@ def predict(file_path='', model_path='15min_dnn_seq'):
     day_lines = read_data(file_path)
     print('数据读取完毕')
 
-    models = []
-    for x in range(0, 12):
-        models.append(load_model(model_path + '_' + str(x) + '.h5'))
-    estimator = joblib.load('km_dmi_18.pkl')
+    model=load_model(model_path + '.h5')
     print('模型加载完毕')
 
     items = sorted(day_lines.keys())
@@ -35,39 +32,25 @@ def predict(file_path='', model_path='15min_dnn_seq'):
 
         up_num = 0
         down_num = 0
-        x = 24 # 每条数据项数
-        k = 18 # 周期
+        size = len(lines[0])
+
         for line in lines:
-            v = line[1:x*k + 1]
-            v = np.array(v)
-            v = v.reshape(k, x)
-            v = v[:,4:8]
-            v = v.reshape(1, 4*k)
-            # print(v)
-            r = estimator.predict(v)
+            train_x = np.array([line[:size - 1]])
+            train_x_a = train_x[:,:18*24]
+            train_x_a = train_x_a.reshape(train_x.shape[0], 18, 24, 1)
+            # train_x_b = train_x[:, 18*18:18*18+2*18]
+            # train_x_b = train_x_b.reshape(train_x.shape[0], 18, 2, 1)
+            train_x_c = train_x[:,18*24:]
 
-            train_x = np.array([line[:-1]])
-            result = models[r[0]].predict(train_x)
+            result = model.predict([train_x_c, train_x_a])
 
             if result[0][3] > 0.5 or result[0][4] > 0.5:
                 down_num = down_num + 1
             elif result[0][1] > 0.5 or result[0][0] > 0.5:
-                up_num = up_num + 0.6  # 乐观调大 悲观调小
-
-            # if result[0][0] > 0.5 or result[0][1] > 0.5:
-            #     if r[0] in [0,2,3,4,5,9,10,11]:
-            #         up_num = up_num + 1
-            #     elif r[0] in [8]:
-            #         up_num = up_num + 0.6
-            #     else:
-            #         up_num = up_num + 0.4
-            # if result[0][3] > 0.5 or result[0][4] > 0.5:
-            #     if r[0] in [4,6,]:
-            #         down_num = down_num + 1
-            #     elif r[0] in [0,1,3,7,8,]:
-            #         down_num = down_num + 0.6
-            #     else:
-            #         down_num = down_num + 0.4
+                up_num = up_num + 1
+            else:
+                up_num = up_num + 0.4 # 乐观调大 悲观调小
+                down_num = down_num + 0.6
 
         print(key, int(up_num), int(down_num), (down_num*1.2 + 2)/(up_num*1.2 + 2))
 
@@ -76,6 +59,6 @@ if __name__ == '__main__':
     # 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\\stock11_18d_20200229.log', model_path='11_18d_dnn_seq')
-    # predict(file_path='D:\\data\\quantization\\stock11_18d_20190103_20190604.log', model_path='14_18d_dnn_seq')
+    # predict(file_path='D:\\data\\quantization\\stock16_18d_20200310.log', model_path='16_18d_mix_seq')
+    predict(file_path='D:\\data\\quantization\\stock16_18d_20191225_20200310_1.log', model_path='16_18d_mix_seq')
     # predict(file_path='D:\\data\\quantization\\stock9_18_4.log', model_path='18d_dnn_seq.h5')

mix/mix_predict_everyday.py

@@ -0,0 +1,189 @@
+# -*- encoding:utf-8 -*-
+import numpy as np
+from keras.models import load_model
+import joblib
+
+
+holder_stock_list = [
+                        '000063.SZ',
+                        '002373.SZ',
+                        '300253.SZ',
+                        '300059.SZ',
+                        '300807.SZ',
+                        '600345.SH',
+                         # b账户
+                        '300422.SZ',
+                        '300468.SZ',
+    ]
+
+
+def read_data(path):
+    lines = []
+    with open(path) as f:
+        for line in f.readlines()[:]:
+            line = eval(line.strip())
+            if line[-2][0].startswith('0') or line[-2][0].startswith('3'):
+                lines.append(line)
+
+    size = len(lines[0])
+    train_x=[s[:size - 2] for s in lines]
+    train_y=[s[size-1] for s in lines]
+    return np.array(train_x),np.array(train_y),lines
+
+
+import pymongo
+from util.mongodb import get_mongo_table_instance
+code_table = get_mongo_table_instance('tushare_code')
+k_table = get_mongo_table_instance('stock_day_k')
+stock_concept_table = get_mongo_table_instance('tushare_concept_detail')
+all_concept_code_list = list(get_mongo_table_instance('tushare_concept').find({}))
+
+
+industry = ['家用电器', '元器件', 'IT设备', '汽车服务',
+            '汽车配件', '软件服务',
+            '互联网', '纺织',
+            '塑料', '半导体',]
+
+A_concept_code_list = [   'TS2', # 5G
+                        'TS24', # OLED
+                        'TS26', #健康中国
+                        'TS43',  #新能源整车
+                        'TS59', # 特斯拉
+                        'TS65', #汽车整车
+                        'TS142', # 物联网
+                        'TS153', # 无人驾驶
+                        'TS163', # 雄安板块-智慧城市
+                        'TS175', # 工业自动化
+                        'TS232', # 新能源汽车
+                        'TS254', # 人工智能
+                        'TS258', # 互联网医疗
+                        'TS264', # 工业互联网
+                        'TS266', # 半导体
+                        'TS269', # 智慧城市
+                        'TS271', # 3D玻璃
+                        'TS295', # 国产芯片
+                        'TS303', # 医疗信息化
+                        'TS323', # 充电桩
+                        'TS328', # 虹膜识别
+                        'TS361', # 病毒
+    ]
+
+
+gainian_map = {}
+hangye_map = {}
+
+def predict_today(day, model='10_18d', log=True):
+    lines = []
+    with open('D:\\data\\quantization\\stock' + model[:-4] + '_' +  str(day) +'.log') as f:
+        for line in f.readlines()[:]:
+            line = eval(line.strip())
+            # if line[-1][0].startswith('0') or line[-1][0].startswith('3'):
+            lines.append(line)
+
+    size = len(lines[0])
+
+    model=load_model(model + '_seq.h5')
+
+    for line in lines:
+        train_x = np.array([line[:size - 1]])
+        train_x_a = train_x[:,:18*24]
+        train_x_a = train_x_a.reshape(train_x.shape[0], 18, 24, 1)
+        # train_x_b = train_x[:, 18*18:18*18+2*18]
+        # train_x_b = train_x_b.reshape(train_x.shape[0], 18, 2, 1)
+        train_x_c = train_x[:,18*24:]
+
+        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 or result[0][1] > 0.5:
+            if line[-1][0].startswith('688'):
+                continue
+            # 去掉ST
+            if stock['name'].startswith('ST') or stock['name'].startswith('N') or stock['name'].startswith('*'):
+                continue
+
+            if stock['ts_code'] in holder_stock_list:
+                print(stock['ts_code'], stock['name'], '维持买入评级')
+
+            # 跌的
+            k_table_list = list(k_table.find({'code':line[-1][0], 'tradeDate':{'$lte':day}}).sort("tradeDate", pymongo.DESCENDING).limit(5))
+            # if k_table_list[0]['close'] > k_table_list[-1]['close']*1.20:
+            #     continue
+            # if k_table_list[0]['close'] < k_table_list[-1]['close']*0.90:
+            #     continue
+            # if k_table_list[-1]['close'] > 80:
+            #     continue
+
+            # 指定某几个行业
+            # if stock['industry'] in industry:
+            concept_code_list = list(stock_concept_table.find({'ts_code':stock['ts_code']}))
+            concept_detail_list = []
+
+            # 处理行业
+            if stock['sw_industry'] in hangye_map:
+                i_c = hangye_map[stock['sw_industry']]
+                hangye_map[stock['sw_industry']] = i_c + 1
+            else:
+                hangye_map[stock['sw_industry']] = 1
+
+            # if len(concept_code_list) > 0:
+            #     for concept in concept_code_list:
+            #         for c in all_concept_code_list:
+            #             if c['code'] == concept['concept_code']:
+            #                 concept_detail_list.append(c['name'])
+            #
+            #                 if c['name'] in gainian_map:
+            #                     g_c = gainian_map[c['name']]
+            #                     gainian_map[c['name']] = g_c + 1
+            #                 else:
+            #                     gainian_map[c['name']] = 1
+
+            print(line[-1], stock['name'], stock['sw_industry'], str(concept_detail_list), 'buy', k_table_list[0]['pct_chg'])
+
+            if log is True:
+                with open('D:\\data\\quantization\\predict\\' + str(day) + '.txt', mode='a', encoding="utf-8") as f:
+                    f.write(str(line[-1]) + ' ' + stock['name'] + ' ' + stock['sw_industry'] + ' ' + str(concept_detail_list) + ' buy' + '\n')
+
+
+            # concept_list = list(stock_concept_table.find({'ts_code':stock['ts_code']}))
+            # concept_list = [c['concept_code'] for c in concept_list]
+
+        elif result[0][2] > 0.5:
+            if stock['ts_code'] in holder_stock_list:
+                print(stock['ts_code'], stock['name'], '震荡评级')
+
+        elif result[0][3] > 0.5 or result[0][4] > 0.5:
+            if stock['ts_code'] in holder_stock_list:
+                print(stock['ts_code'], stock['name'], '赶紧卖出')
+        else:
+            if stock['ts_code'] in holder_stock_list:
+                print(stock['ts_code'], stock['name'], result[0],)
+
+    print(gainian_map)
+    print(hangye_map)
+
+
+def _read_pfile_map(path):
+    s_list = []
+    with open(path, encoding='utf-8') as f:
+        for line in f.readlines()[:]:
+            s_list.append(line)
+    return s_list
+
+
+def join_two_day(a, b):
+    a_list = _read_pfile_map('D:\\data\\quantization\\predict\\' + str(a) + '.txt')
+    b_list = _read_pfile_map('D:\\data\\quantization\\predict\\dmi_' + str(b) + '.txt')
+    for a in a_list:
+        for b in b_list:
+            if a[2:11] == b[2:11]:
+                print(a)
+
+
+if __name__ == '__main__':
+    # predict(file_path='D:\\data\\quantization\\stock6_5_test.log', model_path='5d_dnn_seq.h5')
+    # predict(file_path='D:\\data\\quantization\\stock6_test.log', model_path='15m_dnn_seq.h5')
+    # multi_predict()
+    predict_today(20200310, model='16_18d_mix', log=True)
+    # join_two_day(20200305, 20200305)

mix/mix_train.py

@@ -19,8 +19,8 @@ from keras.callbacks import EarlyStopping
 
 early_stopping = EarlyStopping(monitor='accuracy', patience=5, verbose=2)
 
-epochs= 50
-size = 80000
+epochs= 120
+size = 380000
 
 
 def read_data(path):
@@ -120,8 +120,8 @@ def create_cnn(width, height, depth, filters=32, kernel_size=(5, 6), regress=Fal
 
 # create the MLP and CNN models
 mlp = create_mlp(train_x_c.shape[1], regress=False)
-cnn_0 = create_cnn(18, 16, 1, kernel_size=(3, 5), filters=32, regress=False, output=150)
-cnn_1 = create_cnn(18, 10, 1, kernel_size=(3, 4), filters=32, regress=False, output=120)
+cnn_0 = create_cnn(18, 16, 1, kernel_size=(6, 5), filters=32, regress=False, output=150)
+cnn_1 = create_cnn(18, 10, 1, kernel_size=(6, 6), filters=32, regress=False, output=120)
 
 # create the input to our final set of layers as the *output* of both
 # the MLP and CNN
@@ -176,6 +176,6 @@ print(score)
 print('Test score:', score[0])
 print('Test accuracy:', score[1])
 
-path="19_18d_mix_seq.h5"
+path="16_18d_mix_seq.h5"
 model.save(path)
 model=None

mix/mix_train_1.py

@@ -0,0 +1,173 @@
+import keras
+# -*- encoding:utf-8 -*-
+import numpy as np
+from keras.models import Sequential
+# 优化方法选用Adam(其实可选项有很多，如SGD)
+from keras.optimizers import Adam
+import random
+from keras.models import load_model
+from imblearn.over_sampling import RandomOverSampler
+from keras.utils import np_utils
+# 用于模型初始化，Conv2D模型初始化、Activation激活函数，MaxPooling2D是池化层
+# Flatten作用是将多位输入进行一维化
+# Dense是全连接层
+from keras.layers import Conv2D, Activation, MaxPool2D, Flatten, Dense,Dropout,Input,MaxPooling2D,BatchNormalization,concatenate
+from keras import regularizers
+from keras.models import Model
+
+epochs= 130
+size = 380000
+
+
+def read_data(path):
+    lines = []
+    with open(path) as f:
+        for x in range(size): #380000
+            lines.append(eval(f.readline().strip()))
+
+    random.shuffle(lines)
+    print('读取数据完毕')
+
+    d=int(0.7*len(lines))
+
+    train_x=[s[:-2] for s in lines[0:d]]
+    train_y=[s[-1] for s in lines[0:d]]
+    test_x=[s[:-2] for s in lines[d:]]
+    test_y=[s[-1] for s in lines[d:]]
+
+    print('转换数据完毕')
+
+    ros = RandomOverSampler(random_state=0)
+    X_resampled, y_resampled = ros.fit_sample(np.array(train_x), np.array(train_y))
+
+    print('数据重采样完毕')
+
+    return X_resampled,y_resampled,np.array(test_x),np.array(test_y)
+
+
+train_x,train_y,test_x,test_y=read_data("D:\\data\\quantization\\stock16_18d_train.log")
+
+train_x_a = train_x[:,:18*24]
+train_x_a = train_x_a.reshape(train_x.shape[0], 18, 24, 1)
+# train_x_b = train_x[:, 18*18:18*18+2*18]
+# train_x_b = train_x_b.reshape(train_x.shape[0], 18, 2, 1)
+train_x_c = train_x[:,18*24:]
+
+
+def create_mlp(dim, regress=False):
+    # define our MLP network
+    model = Sequential()
+    model.add(Dense(64, input_dim=dim, activation="relu"))
+    model.add(Dense(64, activation="relu"))
+
+    # check to see if the regression node should be added
+    if regress:
+        model.add(Dense(1, activation="linear"))
+
+    # return our model
+    return model
+
+
+def create_cnn(width, height, depth, filters=(4, 6), kernel_size=(5, 6), regress=False, output=24):
+    # initialize the input shape and channel dimension, assuming
+    # TensorFlow/channels-last ordering
+    inputShape = (width, height, 1)
+    chanDim = -1
+
+    # define the model input
+    inputs = Input(shape=inputShape)
+
+    x = inputs
+
+    # CONV => RELU => BN => POOL
+    x = Conv2D(32, kernel_size, strides=2, padding="same")(x)
+    x = Activation("relu")(x)
+    x = BatchNormalization(axis=chanDim)(x)
+    # x = MaxPooling2D(pool_size=(2, 2))(x)
+    # if width > 2:
+    #     x = Conv2D(32, (10, 6), padding="same")(x)
+    #     x = Activation("relu")(x)
+    #     x = BatchNormalization(axis=chanDim)(x)
+
+    # flatten the volume, then FC => RELU => BN => DROPOUT
+    x = Flatten()(x)
+    x = Dense(output)(x)
+    x = Activation("relu")(x)
+    x = BatchNormalization(axis=chanDim)(x)
+    x = Dropout(0.2)(x)
+
+    # apply another FC layer, this one to match the number of nodes
+    # coming out of the MLP
+    x = Dense(output)(x)
+    x = Activation("relu")(x)
+
+    # check to see if the regression node should be added
+    if regress:
+        x = Dense(1, activation="linear")(x)
+
+    # construct the CNN
+    model = Model(inputs, x)
+
+    # return the CNN
+    return model
+
+
+# create the MLP and CNN models
+mlp = create_mlp(train_x_c.shape[1], regress=False)
+cnn_0 = create_cnn(18, 24, 1, kernel_size=(6, 6), regress=False, output=256)
+# cnn_1 = create_cnn(18, 2, 1, kernel_size=(6,2), regress=False, output=36)
+
+# create the input to our final set of layers as the *output* of both
+# the MLP and CNN
+combinedInput = concatenate([mlp.output, cnn_0.output])
+
+# our final FC layer head will have two dense layers, the final one
+# being our regression head
+x = Dense(512, activation="relu", kernel_regularizer=regularizers.l1(0.003))(combinedInput)
+x = Dropout(0.2)(x)
+x = Dense(512, activation="relu")(x)
+x = Dense(512, activation="relu")(x)
+# 在建设一层
+x = Dense(5, activation="softmax")(x)
+
+# our final model will accept categorical/numerical data on the MLP
+# input and images on the CNN input, outputting a single value (the
+# predicted price of the house)
+model = Model(inputs=[mlp.input, cnn_0.input], outputs=x)
+
+
+print("Starting training ")
+# h = model.fit(train_x, train_y, batch_size=4096*2, epochs=500, shuffle=True)
+
+# compile the model using mean absolute percentage error as our loss,
+# implying that we seek to minimize the absolute percentage difference
+# between our price *predictions* and the *actual prices*
+opt = Adam(lr=1e-3, decay=1e-3 / 200)
+model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=['accuracy'])
+
+# train the model
+print("[INFO] training model...")
+model.fit(
+    [train_x_c, train_x_a], train_y,
+    # validation_data=([testAttrX, testImagesX], testY),
+    # epochs=int(3*train_x_a.shape[0]/1300),
+    epochs=epochs,
+    batch_size=2048, shuffle=True)
+
+test_x_a = test_x[:,:18*24]
+test_x_a = test_x_a.reshape(test_x.shape[0], 18, 24, 1)
+# test_x_b = test_x[:, 18*18:18*18+2*18]
+# test_x_b = test_x_b.reshape(test_x.shape[0], 18, 2, 1)
+test_x_c = test_x[:,18*24:]
+
+# make predictions on the testing data
+print("[INFO] predicting house prices...")
+score  = model.evaluate([test_x_c, test_x_a], test_y)
+
+print(score)
+print('Test score:', score[0])
+print('Test accuracy:', score[1])
+
+path="16_18d_mix_seq.h5"
+model.save(path)
+model=None