在 Pandas、Numpy 和 Tensorflow 中将二维变量转换为一维变量

Question

我正在使用 tensorflow2.1 设计一个股票预测程序。我的predictions 变量在我的程序结束时应该是一个二维变量，但只是一维变量。返回的错误是：

Traceback (most recent call last):
  File "/Users/owner/Desktop/algo/predict.py", line 120, in <module>
    valid['Predictions'] = predictions
  File "/usr/local/lib/python3.7/site-packages/pandas/core/frame.py", line 3487, in __setitem__
    self._set_item(key, value)
  File "/usr/local/lib/python3.7/site-packages/pandas/core/frame.py", line 3563, in _set_item
    self._ensure_valid_index(value)
  File "/usr/local/lib/python3.7/site-packages/pandas/core/frame.py", line 3540, in _ensure_valid_index
    value = Series(value)
  File "/usr/local/lib/python3.7/site-packages/pandas/core/series.py", line 314, in __init__
    data = sanitize_array(data, index, dtype, copy, raise_cast_failure=True)
  File "/usr/local/lib/python3.7/site-packages/pandas/core/internals/construction.py", line 729, in sanitize_array
    raise Exception("Data must be 1-dimensional")
Exception: Data must be 1-dimensional

这是我的代码：

import requests
import math
import numpy as np
import pandas as pd
import pandas_datareader as web
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
from tensorflow.keras.models import Sequential, load_model
from tensorflow.keras.layers import LSTM, Embedding
from tensorflow.keras.layers import Dense
from sklearn.preprocessing import MinMaxScaler
from bs4 import BeautifulSoup
import matplotlib.pyplot as plt
from datetime import datetime, date
plt.style.use('fivethirtyeight')
today = date.today()

df = web.DataReader('GC=F', data_source='yahoo', start='2019-02-14', end=str(today))

#print(df)

plt.figure(figsize=(16,8))
plt.title('GOLD PRICE HISTORY')
plt.plot(df['Close'])
plt.xlabel('Date', fontsize=18)
plt.ylabel('Close price USD($)', fontsize=18)
#plt.show()

#get closing price
data = df.filter(['Close'])

#get closing price values
dataset = data.values

#set training data length to 91% of total data set
training_data_len = math.ceil(len(dataset))
print(training_data_len)

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

#create training data set
#create the scaled training data set
train_data = scaled_data[0: training_data_len , :]

#split the data into x-train and y-train datasets
x_train = []
y_train = []

for i in range(20, len(train_data)):
    x_train.append(train_data[i-20:i, 0])
    y_train.append(train_data[i, 0])

    if i<= 20:

        #print(x_train)
        #print(y_train)
        pass

#convert x-train and y-train to numpy arrays to train models
x_train, y_train = np.array(x_train), np.array(y_train)

#reshape the data, LSTM model expects 3D dataset
x_train = np.reshape(x_train, (x_train.shape[0], x_train.shape[1], 1))

#Build LSTM MODEL
model = tf.keras.Sequential([
    #tf.keras.layers.Embedding(encoder.vocab_size, 64),
    tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(64)),
    tf.keras.layers.Dense(25, activation='relu'),
    tf.keras.layers.Dense(1, activation='sigmoid')
])

model.compile(loss='mean_squared_error',
              optimizer=tf.keras.optimizers.Adam(1e-4),
              metrics=['accuracy'])

model.fit(x_train, y_train, batch_size=1, epochs=1)

#create testing data set
#creat new array containing scaled values
test_data = scaled_data[training_data_len - 20: , :]
print(test_data.shape)

#create the datasets x-test and y-test
x_test=[]
y_test=dataset[training_data_len:, :]
for i in range(20, len(test_data)+1):
    x_test.append(test_data[i-20:i, 0])

#convert data to numpy array
x_test = np.array(x_test)

#reshape data to 3D
x_test = np.reshape(x_test, (x_test.shape[0], x_test.shape[1], 1))
print(x_test.shape)

#Get predicted price values
predictions = model.predict(x_test)
predictions = scaler.inverse_transform(predictions)
print(predictions.shape)


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

#Plot the data
train = data[:training_data_len]
valid = data[training_data_len:]
valid['Predictions'] = predictions

#Visualize the data
plt.figure(figsize=(16,8))
plt.title('Model')
plt.xlabel('Date', fontsize=18)
plt.ylabel('Close Price USD ($)', fontsize=18)
plt.plot(train['Close'])
plt.plot(valid[['Close', 'Predictions']])
plt.legend(['Train', 'Val', 'Predictions'], loc='lower right')
plt.show()

这个错误是由 Pandas、Numpy 还是 Tensorflow 引起的？可以做些什么来满足这个错误陈述？我一直在 GitHub 上寻求帮助，但他们将我重定向到 Stackoverflow。

谢谢。

Answer 1

将valid['Predictions'] = predictions 更改为valid['Predictions'] = np.squeeze(predictions)。错误应该会消失。

Answer 2

您的预测是一个列表列表。

print(predictions) 导致[[1597.7726]]

你可以使用numpy's squeeze方法

...
print(predictions.shape)
predictions = np.squeeze(predictions)
...

或使用简单的解决方法

...
print(predictions.shape)
predictions = predictions[0]
...

解决错误。