RNN Stock Prediction

1. Dataset introduction

     

 Only take the opening data of the third column for forecasting.

2. Environment

        tensorflow2.1、sklearn、pandas、numpy......

3. Implementation and introduction

        

import numpy as np
import tensorflow as tf
from tensorflow.keras.layers import Dropout, Dense, SimpleRNN
import matplotlib.pyplot as plt
import os
import pandas as pd
from sklearn.preprocessing import MinMaxScaler
from sklearn.metrics import mean_squared_error, mean_absolute_error
import math

maotai = pd.read_csv('./SH600519.csv')  # 读取股票文件

training_set = maotai.iloc[0:2426 - 300, 2:3].values  # 前(2426-300=2126)天的开盘价作为训练集,表格从0开始计数，2:3 是提取[2:3)列，前闭后开,故提取出C列开盘价
test_set = maotai.iloc[2426 - 300:, 2:3].values  # 后300天的开盘价作为测试集

# 归一化
sc = MinMaxScaler(feature_range=(0, 1))  # 定义归一化：归一化到(0，1)之间
# 一般先对训练数据fit_transform，对测试数据用transform。
# 对test_data进行fit_transform()就会"overfitting"
# 对test数据而言，我们应该使用训练集得到的平均值和标准差来归一化，而不是test数据集本身的平均值和均方误差
# transform一般会在fit的基础上进行归一化，这里的fit和model.fit是完全不同的概念
# 本例中没有使用fit应该是fit_transform操作已经包含了训练数据集的fit操作并保存了其均值、方差、最大最小值等参数以供transform使用

# fit():计算数据的参数，（均值），（标准差），并存储在对象中（例如实例化的CountVectorizer()等）。
# transform():将这些参数应用到数据集，进行标准化（尺度化）。
# fit_transform()：将前两种方法合并，fit + transform，然后对数据集使用。

training_set_scaled = sc.fit_transform(training_set)  # 求得训练集的最大值，最小值这些训练集固有的属性，并在训练集上进行归一化
test_set = sc.transform(test_set)  # 利用训练集的属性对测试集进行归一化

x_train = []
y_train = []

x_test = []
y_test = []

# 测试集：csv表格中前2426-300=2126天数据
# 利用for循环，遍历整个训练集，提取训练集中连续60天的开盘价作为输入特征x_train，第61天的数据作为标签，for循环共构建2426-300-60=2066组数据。
for i in range(60, len(training_set_scaled)):
    x_train.append(training_set_scaled[i - 60:i, 0])
    y_train.append(training_set_scaled[i, 0])
# 对训练集进行打乱
np.random.seed(7)
np.random.shuffle(x_train)
np.random.seed(7)
np.random.shuffle(y_train)
tf.random.set_seed(7)
# 将训练集由list格式变为array格式
x_train, y_train = np.array(x_train), np.array(y_train)

# 使x_train符合RNN输入要求：[送入样本数， 循环核时间展开步数， 每个时间步输入特征个数]。
# 此处整个数据集送入，送入样本数为x_train.shape[0]即2066组数据；输入60个开盘价，预测出第61天的开盘价，循环核时间展开步数为60; 每个时间步送入的特征是某一天的开盘价，只有1个数据，故每个时间步输入特征个数为1
x_train = np.reshape(x_train, (x_train.shape[0], 60, 1))
# 测试集：csv表格中后300天数据
# 利用for循环，遍历整个测试集，提取测试集中连续60天的开盘价作为输入特征x_train，第61天的数据作为标签，for循环共构建300-60=240组数据。
for i in range(60, len(test_set)):
    x_test.append(test_set[i - 60:i, 0])
    y_test.append(test_set[i, 0])
# 测试集变array并reshape为符合RNN输入要求：[送入样本数， 循环核时间展开步数， 每个时间步输入特征个数]
x_test, y_test = np.array(x_test), np.array(y_test)
x_test = np.reshape(x_test, (x_test.shape[0], 60, 1))

model = tf.keras.Sequential([
    SimpleRNN(80, return_sequences=True),
    Dropout(0.2),
    SimpleRNN(100),
    Dropout(0.2),
    Dense(1)
])

model.compile(optimizer=tf.keras.optimizers.Adam(0.001),
              loss='mean_squared_error')  # 损失函数用均方误差
# 该应用只观测loss数值，不观测准确率，所以删去metrics选项，一会在每个epoch迭代显示时只显示loss值

checkpoint_save_path = "./checkpoint/rnn_stock.ckpt"

if os.path.exists(checkpoint_save_path + '.index'):
    print('-------------load the model-----------------')
    model.load_weights(checkpoint_save_path)

cp_callback = tf.keras.callbacks.ModelCheckpoint(filepath=checkpoint_save_path,
                                                 save_weights_only=True,
                                                 save_best_only=True,
                                                 monitor='val_loss')

history = model.fit(x_train, y_train, batch_size=64, epochs=50, validation_data=(x_test, y_test), validation_freq=1,
                    callbacks=[cp_callback])

model.summary()

file = open('./weights.txt', 'w')  # 参数提取
for v in model.trainable_variables:
    file.write(str(v.name) + '\n')
    file.write(str(v.shape) + '\n')
    file.write(str(v.numpy()) + '\n')
file.close()

loss = history.history['loss']
val_loss = history.history['val_loss']

plt.plot(loss, label='Training Loss')
plt.plot(val_loss, label='Validation Loss')
plt.title('Training and Validation Loss')
plt.legend()
plt.show()

################## predict ######################
# 测试集输入模型进行预测
predicted_stock_price = model.predict(x_test)
# 对预测数据还原---从（0，1）反归一化到原始范围
predicted_stock_price = sc.inverse_transform(predicted_stock_price)
# 对真实数据还原---从（0，1）反归一化到原始范围
real_stock_price = sc.inverse_transform(test_set[60:])
# 画出真实数据和预测数据的对比曲线
plt.plot(real_stock_price, color='red', label='MaoTai Stock Price')
plt.plot(predicted_stock_price, color='blue', label='Predicted MaoTai Stock Price')
plt.title('MaoTai Stock Price Prediction')
plt.xlabel('Time')
plt.ylabel('MaoTai Stock Price')
plt.legend()
plt.show()

##########evaluate##############
# calculate MSE 均方误差 ---> E[(预测值-真实值)^2] (预测值减真实值求平方后求均值)
mse = mean_squared_error(predicted_stock_price, real_stock_price)
# calculate RMSE 均方根误差--->sqrt[MSE]    (对均方误差开方)
rmse = math.sqrt(mean_squared_error(predicted_stock_price, real_stock_price))
# calculate MAE 平均绝对误差----->E[|预测值-真实值|](预测值减真实值求绝对值后求均值）
mae = mean_absolute_error(predicted_stock_price, real_stock_price)
print('均方误差: %.6f' % mse)
print('均方根误差: %.6f' % rmse)
print('平均绝对误差: %.6f' % mae)

Results of the:

loss curve:

Forecast renderings:

Overall result:

First, slice the data in csv --> iloc function

Then, normalize the data, pay attention here:

# Generally, fit_transform is used for training data first, and transform is used for test data.
# Fit_transform() on test_data will be "overfitting"
# Reason for overfitting: For test data, we should use the mean and standard deviation obtained from the training set to normalize, not the mean of the test data set itself And mean square error
# transform will generally be normalized on the basis of fit, note that fit and model.fit here are completely different concepts
# fit is not used in this example, it should be that the fit_transform operation already includes the training data set fit operates and saves its mean, variance, maximum and minimum values ​​and other parameters for use by transform

The difference between these functions:

# fit(): Calculate the parameters of the data, (mean value), (standard deviation), and store them in the object (such as instantiated CountVectorizer(), etc.).
# transform(): Apply these parameters to the data set for normalization (scaling).
# fit_transform(): Combine the first two methods, fit + transform, and then use it on the dataset.
 

Finally, the normalized content can be restored by the function  inverse_transform function 

Randomly shuffle data and labels, perform iterative training, save the model, extract parameters and save, visualize data, predict model accuracy and visualize.

Reference link:

https://blog.csdn.net/yyhhlancelot/article/details/85097656?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-2.nonecase&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-2.nonecase

https://blog.csdn.net/weixin_39190382/article/details/106383101