在运行以下代码时出现错误发现数组为dim3估计量应为<= 2

如何解决在运行以下代码时出现错误发现数组为dim3估计量应为<= 2

检查模型的鲁棒性： 在本节中，我将检查我的LSTM模型的鲁棒性。从2017年7月1日到2017年7月20日，我为此使用了新的看不见的数据集。我已经从Google财经网站下载了数据集，以检查模型的稳健性。

import preprocess_data as ppd
data = pd.read_csv('E:/DBSOM DATA\FOM_Sem 2/Analyses of S&U Data/Project work/Stock-Price-Prediction-master/googl.csv')
stocks = ppd.remove_data(data)
stocks = ppd.get_normalised_data(stocks)
stocks = stocks.drop(['Item'],axis = 1)

#Print the data frame head and tail

print(stocks.head())
X = stocks[:].values
Y = stocks[:]['Close'].values
X = sd.unroll(X,1)
Y = Y[-X.shape[0]:]

print(X.shape)
print(Y.shape)

# Generate predictions 

predictions = model.predict(X)

#get the test score

testScore = model.evaluate(X,Y,verbose=0)
print('Test Score: %.4f MSE (%.4f RMSE)' % (testScore,math.sqrt(testScore)))

功能定义

import pandas as pd
Import sklearn.preprocessing.StandardScaler
from sklearn.preprocessing import MinMaxScaler

def get_normalised_data(data):
    
    # Initialize a scaler,then apply it to the features
    
    scaler = MinMaxScaler()
    numerical = ['Open','Close','Volume']
    data[numerical] = scaler.fit_transform(data[numerical])
    return data

def remove_data(data):

    # Define columns of data to keep from historical stock data

    item = []
    open = []
    close = []
    volume = []

    # Loop through the stock data objects backwards and store factors we want to keep

    i_counter = 0
    for i in range(len(data) - 1,-1,-1):
        item.append(i_counter)
        open.append(data['Open'][i])
        close.append(data['Close'][i])
        volume.append(data['Volume'][i])
        i_counter += 1

    # Create a data frame for stock data
    stocks = pd.DataFrame()

    # Add factors to data frame

    stocks['Item'] = item
    stocks['Open'] = open
    stocks['Close'] = pd.to_numeric(close)
    stocks['Volume'] = pd.to_numeric(volume)

    # return new formatted data
    return stocks

我已经花了很多时间解决此错误，但是没有找到解决方案。请提出建议。

解决方法

我删除了此行Import sklearn.preprocessing.StandardScaler

和沃尔拉。一切运行顺利。