数据读取问题

我按照下面例题读取数据可是到了“2. Set last day Adjusted Close as y”这步，数据总是缺少最后一行。不知道为什么求高人解答。读取的源数据df可以看到最后一行值都是1.00000，可是转换到y_test之后最后一行就不见了。


代码如下：

1. import numpy as np
   
2. import matplotlib.pyplot as plt
   
3. import matplotlib.pyplot as plt2
   
4. import pandas as pd
   
5. from pandas import datetime
   
6. import math, time
   
7. import itertools
   
8. from sklearn import preprocessing
   
9. import datetime
   
10. from sklearn.metrics import mean_squared_error
    
11. from math import sqrt
    
12. from keras.models import Sequential
    
13. from keras.layers.core import Dense, Dropout, Activation
    
14. from keras.layers.recurrent import LSTM
    
15. from keras.models import load_model
    
16. import keras
    
17. import pandas_datareader.data as web
    
18. import h5py

复制代码

1. def get_stock_data(stock_name, normalize=True):
   
2.     start = datetime.datetime(1971, 1, 1)
   
3.     end = datetime.date.today()
   
4.     df = web.DataReader(stock_name, "yahoo", start, end)
   
5.     df.drop(['Volume', 'Close'], 1, inplace=True)
   
6. 
   
7.     if normalize:        
   
8.         min_max_scaler = preprocessing.MinMaxScaler()
   
9.         df['Open'] = min_max_scaler.fit_transform(df.Open.values.reshape(-1,1))
   
10.         df['High'] = min_max_scaler.fit_transform(df.High.values.reshape(-1,1))
    
11.         df['Low'] = min_max_scaler.fit_transform(df.Low.values.reshape(-1,1))
    
12.         df['Adj Close'] = min_max_scaler.fit_transform(df['Adj Close'].values.reshape(-1,1))
    
13.     return df
    
14. 
    
15. 
    
16. df = get_stock_data(stock_name, normalize=True)
    
17. 
    
18. df

复制代码

2. Set last day Adjusted Close as y

1. def load_data(stock, seq_len):
   
2.     amount_of_features = len(stock.columns)
   
3.     data = stock.as_matrix()
   
4.     sequence_length = seq_len + 1 # index starting from 0
   
5.     result = []
   
6. 
   
7.     for index in range(len(data) - sequence_length): # maxmimum date = lastest date - sequence length
   
8.         result.append(data[index: index + sequence_length]) # index : index + 22days
   
9. 
   
10.     result = np.array(result)
    
11.     row = round(0.9 * result.shape[0]) # 90% split
    
12. 
    
13.     train = result[:int(row), :] # 90% date
    
14.     X_train = train[:, :-1] # all data until day m
    
15.     y_train = train[:, -1][:,-1] # day m + 1 adjusted close price
    
16. 
    
17.     X_test = result[int(row):, :-1]
    
18.     y_test = result[int(row):, -1][:,-1]
    
19. 
    
20.     X_train = np.reshape(X_train, (X_train.shape[0], X_train.shape[1], amount_of_features))
    
21.     X_test = np.reshape(X_test, (X_test.shape[0], X_test.shape[1], amount_of_features))  
    
22. 
    
23.     return [X_train, y_train, X_test, y_test]
    
24. 
    
25. y_test

复制代码

您好，如果您的求助没有解决，请到项目交易发布需求，会有更快更专业的用户帮助您 https://bbs.pinggu.org/z_prj.php