本文是对：

https://machinelearningmastery.com/time-series-forecasting-long-short-term-memory-network-python/

https://blog.csdn.net/iyangdi/article/details/77868744

两篇博文的学习笔记，两个博主笔风都很浪，有些细节一笔带过，本人以谦逊的态度进行了学习和整理，笔记内容都在代码的注释中。有不清楚的可以去原博主文中查看。

数据集下载：https://datamarket.com/data/set/22r0/sales-of-shampoo-over-a-three-year-period

后期我会补上我的github

这一节对第四节的代码进行了升级，进行了30次重复预测，并且对最后得到的RMSE的值求了方差，从而达到对预测性能进行一个评估股的作用
 

# coding=utf-8
from pandas import read_csv
from pandas import datetime
from pandas import concat
from pandas import DataFrame
from pandas import Series
from sklearn.metrics import mean_squared_error
from sklearn.preprocessing import MinMaxScaler
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import LSTM
from math import sqrt
from matplotlib import pyplot
import numpy


# 读取时间数据的格式化
def parser(x):
    return datetime.strptime(x, '%Y/%m/%d')


# 转换成有监督数据
def timeseries_to_supervised(data, lag=1):
    df = DataFrame(data)
    columns = [df.shift(i) for i in range(1, lag + 1)]  # 数据滑动一格，作为input，df原数据为output
    columns.append(df)
    df = concat(columns, axis=1)
    df.fillna(0, inplace=True)
    return df


# 转换成差分数据
def difference(dataset, interval=1):
    diff = list()
    for i in range(interval, len(dataset)):
        value = dataset[i] - dataset[i - interval]
        diff.append(value)
    return Series(diff)


# 逆差分
def inverse_difference(history, yhat, interval=1):  # 历史数据，预测数据，差分间隔
    return yhat + history[-interval]


# 缩放
def scale(train, test):
    # 根据训练数据建立缩放器
    scaler = MinMaxScaler(feature_range=(-1, 1))
    scaler = scaler.fit(train)
    # 转换train data
    train = train.reshape(train.shape[0], train.shape[1])
    train_scaled = scaler.transform(train)
    # 转换test data
    test = test.reshape(test.shape[0], test.shape[1])
    test_scaled = scaler.transform(test)
    return scaler, train_scaled, test_scaled


# 逆缩放
def invert_scale(scaler, X, value):
    new_row = [x for x in X] + [value]
    array = numpy.array(new_row)
    array = array.reshape(1, len(array))
    inverted = scaler.inverse_transform(array)
    return inverted[0, -1]


# fit LSTM来训练数据
def fit_lstm(train, batch_size, nb_epoch, neurons):
    X, y = train[:, 0:-1], train[:, -1]
    X = X.reshape(X.shape[0], 1, X.shape[1])
    model = Sequential()
    # 添加LSTM层
    model.add(LSTM(neurons, batch_input_shape=(batch_size, X.shape[1], X.shape[2]), stateful=True))
    model.add(Dense(1))  # 输出层1个node
    # 编译，损失函数mse+优化算法adam
    model.compile(loss='mean_squared_error', optimizer='adam')
    for i in range(nb_epoch):
        # 按照batch_size，一次读取batch_size个数据
        model.fit(X, y, epochs=1, batch_size=batch_size, verbose=0, shuffle=False)
        model.reset_states()
        print("当前计算次数："+str(i))
    return model


# 1步长预测
def forcast_lstm(model, batch_size, X):
    X = X.reshape(1, 1, len(X))
    yhat = model.predict(X, batch_size=batch_size)
    return yhat[0, 0]


# 加载数据
series = read_csv('data_set/shampoo-sales.csv', header=0, parse_dates=[0], index_col=0, squeeze=True,
                  date_parser=parser)

# 让数据变成稳定的
raw_values = series.values
diff_values = difference(raw_values, 1)#转换成差分数据

# 把稳定的数据变成有监督数据
supervised = timeseries_to_supervised(diff_values, 1)
supervised_values = supervised.values

# 数据拆分：训练数据、测试数据，前24行是训练集，后12行是测试集
train, test = supervised_values[0:-12], supervised_values[-12:]

# 数据缩放
scaler, train_scaled, test_scaled = scale(train, test)



#重复实验   
repeats = 30
error_scores = list()
for r in range(repeats):
    # fit 模型
    lstm_model = fit_lstm(train_scaled, 1, 100, 4)  # 训练数据，batch_size，epoche次数, 神经元个数
    # 预测
    train_reshaped = train_scaled[:, 0].reshape(len(train_scaled), 1, 1)#训练数据集转换为可输入的矩阵
    lstm_model.predict(train_reshaped, batch_size=1)#用模型对训练数据矩阵进行预测
    # 测试数据的前向验证，实验发现，如果训练次数很少的话，模型回简单的把数据后移，以昨天的数据作为今天的预测值，当训练次数足够多的时候
    # 才会体现出来训练结果
    predictions = list()
    for i in range(len(test_scaled)):
        # 1步长预测
        X, y = test_scaled[i, 0:-1], test_scaled[i, -1]
        yhat = forcast_lstm(lstm_model, 1, X)
        # 逆缩放
        yhat = invert_scale(scaler, X, yhat)
        # 逆差分
        yhat = inverse_difference(raw_values, yhat, len(test_scaled) + 1 - i)
        predictions.append(yhat)
        expected = raw_values[len(train) + i + 1]
        print('Moth=%d, Predicted=%f, Expected=%f' % (i + 1, yhat, expected))
    # 性能报告
    rmse = sqrt(mean_squared_error(raw_values[-12:], predictions))
    print('%d) Test RMSE:%.3f' %(r+1,rmse))
    error_scores.append(rmse)

#统计信息   
results = DataFrame()
results['rmse'] = error_scores
print(results.describe())
results.boxplot()
pyplot.show()