使用 LSTM 将单变量转换为多变量时间序列预测

Question

I am new to the world of artificial neural networks so if I make some mistakes, excuse me and correct me if you can. I would like to use an LSTM model to be able to predict the price of bitcoin in the market. I know the practical limitations of the model but I am creating it for educational purposes.

I don't know whether to define it a multilayer or multivariate model (if someone could explain the difference I would be grateful for) basically a model that trained on the closing price called 'close' can predict the closing price of the next day by observing the previous 60 days.

I had no problems building the model from here I just spoke to you, the problem is that I would like to train the model with other information such as the volume or the maximum price of the day. The important thing is to be able to decide which two types of information to insert in the model. I found a site where the Multivariate Time Series Forecasting with LSTMs in Keras is explained in detail but I cannot apply it to my specific case. Could you help me integrate the 'volume' variable into the model to see if the predictive power of the future 'close' closing price improves or worsens?

The data are of this type and can be downloaded here from kaggle --> Download

import pandas as pd
import math
import numpy as np
from sklearn.preprocessing import MinMaxScaler
from keras.models import Sequential
from keras.layers import Dense, LSTM
import matplotlib.pyplot as plt


#Create a new dataframe with only the 'Close column
data = df.filter(['close'])
#Convert the dataframe to a numpy array
dataset = data.values
#Get the number of rows to train the model on
training_data_len = math.ceil( len(dataset) * .8 )

#scale data
scaler = MinMaxScaler(feature_range=(0, 1))
scaled_data = scaler.fit_transform(dataset)

#Create the scaled training data set
train_data = scaled_data[0:training_data_len , :]
#Split the data into x_train and y_train data sets
x_train = []
y_train = []

for i in range(60, len(train_data)):
  x_train.append(train_data[i-60:i, 0])
  y_train.append(train_data[i, 0])
  # if i<= 61:
    # print(x_train)
    # print(y_train)
    # print()

#Convert the x_train and y_train to numpy arrays
x_train, y_train = np.array(x_train), np.array(y_train)

#Reshape the data
x_train = np.reshape(x_train, (x_train.shape[0], x_train.shape[1], 1))
print (x_train.shape)

#Build the LSTM model
model = Sequential()
model.add(LSTM(50, return_sequences=True, input_shape= (x_train.shape[1], 1)))
model.add(LSTM(50, return_sequences= False))
model.add(Dense(25))
model.add(Dense(1))

#Compile the model
model.compile(optimizer='adam', loss='mean_squared_error')
#Train the model
model.fit(x_train, y_train, batch_size=1, epochs=1)

#Create the testing data set
#Create a new array containing scaled values 
test_data = scaled_data[training_data_len - 60: , :]
#Create the data sets x_test and y_test
x_test = []
y_test = dataset[training_data_len:, :]
for i in range(60, len(test_data)):
  x_test.append(test_data[i-60:i, 0])

#Convert the data to a numpy array
x_test = np.array(x_test)
#Reshape the data

x_test = np.reshape(x_test, (x_test.shape[0], x_test.shape[1], 1 ))

# print (len(x_test))
# #Get the models predicted price values
predictions = model.predict(x_test)

predictions = scaler.inverse_transform(predictions)
print(predictions)

#Get the root mean squared error (RMSE)
rmse=np.sqrt(np.mean(((predictions- y_test)**2)))
print (rmse)

Answer 1

From the code and from the Comments, I understand that you are performing Time Series forecasting for Uni-Variate Data (with only column being Close ) and now, want to perform Time Series Forecasting for Multi-Variate Data (with the Columns, Close and Volume ).

Important part of code for you will be the function, multivariate_data , which returns the Features and Labels according to the past history of 60 days, and a Target Date of 1 Day.

past_history = 60
future_target = 1

Complete working code (till Training) for Multi_Variate Data is shown below:

import pandas as pd
import math
import numpy as np
from sklearn.preprocessing import MinMaxScaler
from keras.models import Sequential
from keras.layers import Dense, LSTM
import matplotlib.pyplot as plt

df = pd.read_csv('datasets_101543_240726.csv')

#Create a new dataframe with only the 'Close column
data = df.filter(['close', 'Volume'])
#Convert the dataframe to a numpy array
dataset = data.values
#Get the number of rows to train the model on
training_data_len = math.ceil( len(dataset) * .8 )

# scale data
data_mean = dataset[:training_data_len].mean(axis=0)
data_std = dataset[:training_data_len].std(axis=0)
dataset = (dataset-data_mean)/data_std

def multivariate_data(dataset, target, start_index, end_index, history_size,
                      target_size):
  data = []
  labels = []

  start_index = start_index + history_size
  if end_index is None:
    end_index = len(dataset) - target_size

  for i in range(start_index, end_index):
    indices = range(i-history_size, i)
    data.append(dataset[indices])

    labels.append(target[i:i+target_size])

  return np.array(data), np.array(labels)

past_history = 60
future_target = 1

x_train, y_train = multivariate_data(dataset, dataset[:, 0], 0,
                                                   training_data_len, past_history,
                                                   future_target)
x_val, y_val = multivariate_data(dataset, dataset[:, 0],
                                               training_data_len, None, past_history,
                                               future_target)


#Reshape the data
#x_train = np.reshape(x_train, (x_train.shape[0], x_train.shape[1], 1))
#print (x_train.shape)

#Build the LSTM model
model = Sequential()
model.add(LSTM(50, return_sequences=True, input_shape= (x_train.shape[1], 1)))
model.add(LSTM(50, return_sequences= False))
model.add(Dense(25))
model.add(Dense(1))

#Compile the model
model.compile(optimizer='adam', loss='mean_squared_error')
#Train the model
model.fit(x_train, y_train, batch_size=1, epochs=1)

Test Data in your code, x_test and y_test can be replaced with x_val and y_val and you can perform predictions on that data.

Please refer Tensorflow Tutorial related to Time Series Forecasting on Multi Variate Data for complete code.

Hope this helps. Happy Learning!