气温预测--单特征

一、了解

数据集：气温数据，多种指标

任务目标：预测未来某一时间点的气温/未来某一时间片段的气温

数据集下载链接如下：jena_climate_2009_2016

二、实战--单特征

第一步，加载数据集

df = pd.read_csv('./datasets/jena_climate_2009_2016.csv')
df.head()

此数据集的数据，是每个十分钟一条数据，每个数据有多个指标

第二步，构建序列数据函数

'''
dataset表示输入的数据集，里面可以是一个或者多个列特征

history_size表示时间窗口的大小

indices = range（i-history_size，i）表示窗口序列索引，i表示每个窗口的起始位置

target_size 表示要预测的结果是窗口的第几个时间点，0则表示下一个时间点的预测结果，取其当做标签
'''
def univariate_data(dataset, 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)
        #Reshape data from (history_size,) to (history_size,1)
        data.append(np.reshape(dataset[indices], (history_size, 1)))
        labels.append(dataset[i+target_size])
    return np.array(data), np.array(labels)

第三步，取出数据DF

#前30W个样本数据当训练集，剩下的当验证集
TRAIN_SPLIT = 300000

#取出单特征数据df， 只选一个温度特征
uni_data = df['T (degC)']
uni_data.index = df['Date Time']
uni_data.head()

#展示当前特征
uni_data.plot(subplots=True)

#得到温度数据
uni_data = uni_data.values
print( uni_data.shape )

#数据预处理，归一化
uni_train_mean = uni_data[:TRAIN_SPLIT].mean()
uni_train_std = uni_data[:TRAIN_SPLIT].std()
uni_data = (uni_data - uni_train_mean)/uni_train_std

#窗口大小为20条数据，预测一个时刻的气温
univariate_past_history = 20
univariate_future_target = 0
x_train_uni, y_train_uni = univariate_data(uni_data, 0, TRAIN_SPLIT,
                           univariate_past_history,
                           univariate_future_target)
x_val_uni, y_val_uni = univariate_data(uni_data, TRAIN_SPLIT,
                                      None, univariate_past_history,
                                      univariate_future_target)

#维度
print( x_train_uni.shape )
print( y_train_uni.shape )
print( x_val_uni.shape )
print( y_val_uni.shape )

第四步，时间序列数据展示

# 创建 以当前天为0，之前的 20天[-20,-19,,,,-1]
def create_time_steps(length):
    time_steps = []
    for i in range(-length, 0, 1):
        time_steps.append(i)
    return time_steps

def show_plot(plot_data, delta, title):
    labels = ['History', 'True Future', 'Model Prediction']
    marker = ['.-', 'rx', 'go']
    time_steps = create_time_steps(plot_data[0].shape[0])
    if delta:
        future = delta
    else:
        future = 0
        
    plt.title(title)
    for i, x in enumerate(plot_data):
        if i:
            plt.plot(future, plot_data[i], marker[i],
                    markersize=10, label=labels[i])
        else:
            plt.plot(time_steps, plot_data[i].flatten(),
                    marker[i], label=labels[i])
    plt.legend()
    plt.xlim([time_steps[0], (future+5)*2])
    plt.xlabel('Time-Step')
    return plt

show_plot([x_train_uni[0], y_train_uni[0]], 0, 'Sample Example')

第五步，用from_tensor_slices创建Tensor类型训练数据和验证数据

BATCH_SIZE = 256
BUFFER_SIZE = 10000

train_univariate = tf.data.Dataset.from_tensor_slices((x_train_uni, y_train_uni))
train_univariate = train_univariate.cache().shuffle(BUFFER_SIZE).batch(BATCH_SIZE).repeat()

val_univariate = tf.data.Dataset.from_tensor_slices((x_val_uni, y_val_uni))
val_univariate = val_univariate.batch(BATCH_SIZE).repeat()

第六步，用 LSTM模型 预测气温

simple_lstm_model = tf.keras.models.Sequential([
    keras.layers.LSTM(8, input_shape=x_train_uni.shape[-2:]),  #输入维度为 (20,1)
    keras.layers.Dense(1)
])

simple_lstm_model.compile(optimizer=tf.optimizers.Adam(),
                         loss='mae')

simple_lstm_model.summary()

运行结果如下：

Model: "sequential_1"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
lstm_1 (LSTM)                (None, 8)                 320       
_________________________________________________________________
dense_1 (Dense)              (None, 1)                 9         
=================================================================
Total params: 329
Trainable params: 329
Non-trainable params: 0

第七步，先预测一下测试结果

#得到一个batch的测试结果
for x, y in val_univariate.take(1):
    print(simple_lstm_model.predict(x).shape)

第八步，训练模型

# 为了训练能快点，一个epoch就只训练200次
EVALUATION_INTERVAL = 200
EPOCHS = 10

simple_lstm_model.fit(train_univariate, 
                     epochs=EPOCHS,
                     steps_per_epoch=EVALUATION_INTERVAL,
                     validation_data=val_univariate,
                     validation_steps=50)

第九步，用训练好的模型预测结果

#展示其中部分数据的预测结果
# x 是 256个样例，即一个batch
for x, y in val_univariate.take(3):
    plot = show_plot([x[0].numpy(), y[0].numpy(),
                     simple_lstm_model.predict(x)[0]],
                    0, 'Simple LSTM model')
    plot.show()

后面，将用多特征来训练网络模型