LSTM

guowen1979 · May 11, 2021 · f4b68ef · f4b68ef
1 parent d1e95d9
commit f4b68ef
Show file tree

Hide file tree

Showing 55 changed files with 6,056 additions and 0 deletions.
diff --git a/.idea/.gitignore b/.idea/.gitignore
diff --git a/.idea/codeStyles/Project.xml b/.idea/codeStyles/Project.xml
diff --git a/.idea/codeStyles/codeStyleConfig.xml b/.idea/codeStyles/codeStyleConfig.xml
diff --git a/.idea/inspectionProfiles/profiles_settings.xml b/.idea/inspectionProfiles/profiles_settings.xml
diff --git a/.idea/misc.xml b/.idea/misc.xml
diff --git a/.idea/modules.xml b/.idea/modules.xml
diff --git a/.idea/ticket.iml b/.idea/ticket.iml
diff --git a/LSTM_pre.py b/LSTM_pre.py
@@ -0,0 +1,179 @@
+# -*- coding: utf-8 -*-
+import pandas as pd
+import numpy as np
+import os,sys
+import time
+import tensorflow as tf
+import matplotlib.pyplot as plt
+
+path = "./data.csv"
+
+# time_step = 24  # 时间步
+# rnn_unit = 10  # hidden layer units
+# batch_size = 60  # 每一批次训练多少个样例
+# input_size = 1  # 输入层维度
+# output_size = 1  # 输出层维度
+# lr = 0.0006  # 学习率
+#
+# batch =1
+
+def LSTM_test(normalize_data,statusname):
+    #设置常量
+    time_step=24      #时间步
+    rnn_unit=10       #hidden layer units
+    batch_size=60     #每一批次训练多少个样例
+    input_size=1      #输入层维度
+    output_size=1     #输出层维度
+    lr=0.0006         #学习率
+    train_x,train_y=[],[]   #训练集
+    for i in range(len(normalize_data)-time_step-1):
+        x=normalize_data[i:i+time_step]
+        y=normalize_data[i+1:i+time_step+1]
+        train_x.append(x.tolist())
+        train_y.append(y.tolist())
+
+
+
+    X=tf.placeholder(tf.float32, [None,time_step,input_size])    #每批次输入网络的tensor
+    Y=tf.placeholder(tf.float32, [None,time_step,output_size]) #每批次tensor对应的标签
+
+    #输入层、输出层权重、偏置
+    weights={
+        'in':tf.Variable(tf.random_normal([input_size,rnn_unit])),
+        'out':tf.Variable(tf.random_normal([rnn_unit,1]))
+    }
+    biases={
+        'in':tf.Variable(tf.constant(0.1,shape=[rnn_unit,])),
+        'out':tf.Variable(tf.constant(0.1,shape=[1,]))
+    }
+
+
+
+    def lstm(batch):  #参数：输入网络批次数目
+        w_in=weights['in']
+        b_in=biases['in']
+        input=tf.reshape(X,[-1,input_size])  #需要将tensor转成2维进行计算，计算后的结果作为隐藏层的输入
+        input_rnn=tf.matmul(input,w_in)+b_in
+        input_rnn=tf.reshape(input_rnn,[-1,time_step,rnn_unit])  #将tensor转成3维，作为lstm cell的输入
+        cell=tf.nn.rnn_cell.BasicLSTMCell(rnn_unit)
+        init_state=cell.zero_state(batch,dtype=tf.float32)
+        output_rnn,final_states=tf.nn.dynamic_rnn(cell, input_rnn,initial_state=init_state, dtype=tf.float32)  #output_rnn是记录lstm每个输出节点的结果，final_states是最后一个cell的结果
+        output=tf.reshape(output_rnn,[-1,rnn_unit]) #作为输出层的输入
+        w_out=weights['out']
+        b_out=biases['out']
+        pred=tf.matmul(output,w_out)+b_out
+        return pred,final_states
+
+
+    def train_lstm():
+        global batch_size
+        pred,_=lstm(batch_size)
+        #损失函数
+        loss=tf.reduce_mean(tf.square(tf.reshape(pred,[-1])-tf.reshape(Y, [-1])))
+        train_op=tf.train.AdamOptimizer(lr).minimize(loss)
+        saver=tf.train.Saver(tf.global_variables())
+        with tf.Session() as sess:
+            sess.run(tf.global_variables_initializer())
+            #重复训练10000次
+            for i in range(300):
+                step=0
+                start=0
+                end=start+batch_size
+                while(end<len(train_x)):
+                    _,loss_=sess.run([train_op,loss],feed_dict={X:train_x[start:end],Y:train_y[start:end]})
+                    start+=batch_size
+                    end=start+batch_size
+                    #每10步保存一次参数
+                    if step%10==0:
+                        print(i,step,loss_)
+                        print("保存模型：",saver.save(sess,'./model_step_%s/stock.ckpt'%statusname))
+                    step+=1
+
+
+    def prediction():
+        pred,_=lstm(1)    #预测时只输入[1,time_step,input_size]的测试数据
+        saver=tf.train.Saver(tf.global_variables())
+        with tf.Session() as sess:
+            #参数恢复
+            module_file = tf.train.latest_checkpoint('./model_step_%s/'%statusname)
+            saver.restore(sess, module_file)
+            #取训练集最后一行为测试样本。shape=[1,time_step,input_size]
+            prev_seq=train_x[-1]
+            next_seq=sess.run(pred,feed_dict={X:[prev_seq]})
+            data = next_seq[-1]
+            print("预测结果(归一化)：",data)
+        return data
+
+
+    with tf.variable_scope('train'):
+        data = prediction()
+    tf.reset_default_graph()
+    return data
+
+
+
+
+
+
+
+def pre_last(status ="测试"):
+    print(status)
+    if status == "测试":
+        print("请重新输入参数：（具体是哪个球？）")
+        os._exit(0)
+    elif status =="红1":
+        statusname = "red1"
+    elif status == "红2":
+        statusname = "red2"
+    elif status == "红3":
+        statusname = "red3"
+    elif status == "红4":
+        statusname = "red4"
+    elif status == "红5":
+        statusname = "red5"
+    elif status == "红6":
+        statusname = "red6"
+    elif status == "蓝":
+        statusname = "blue"
+    else:
+        print("请重新输入参数：（具体是哪个球？）")
+        os._exit(0)
+    data = pd.read_csv(path,header=[0])
+    dt = np.array(data['%s'%status])
+
+
+    dt_last = dt[-24:]
+    mean_std = pd.read_csv("./model_step/std_mean_%s.csv"%(statusname))
+    dt_std = np.array(mean_std["data_std"])
+    dt_mean = np.array(mean_std["data_mean"])
+    normalize_data = (dt - dt_mean)/dt_std
+    normalize_data = normalize_data[:,np.newaxis]
+
+    #
+    data_pre = LSTM_test(normalize_data,statusname)
+    data_pre = data_pre*dt_std + dt_mean
+    data_pre = float(data_pre)
+    print("预测结果（实际）：",data_pre)
+    return data_pre,dt_std,dt_mean
+
+
+
+
+
+
+if __name__=="__main__":
+    data_new =[]
+    data_pre = []
+    for i in range(1,8,1):
+        if i ==7:
+            status = "蓝"
+        else:
+            status = "红%s"%i
+        print(i)
+        data,dt_std,dt_mean = pre_last(status)
+        data_new.append(data)
+    data_pre.append((data_new))
+    datas = pd.DataFrame(data_pre,columns=["红1","红2","红3","红4","红5","红6","蓝"])
+    datas.to_csv("./data_pre.csv",index=False)
+
+