python爬虫将数据写入csv以及LSTM序列预测

置顶 已于 2022-03-03 09:23:49 修改
阅读量9.6k 收藏 50
点赞数 8
分类专栏： 爬虫 python 时间序列预测 文章标签： python lstm 爬虫
于 2020-06-30 09:49:51 首次发布
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/brytlevson/article/details/107034278
版权

python爬虫并且将数据写入csv的三种方法，前面两种是pandas写入csv ,后面是with open/open 直接写入，直接上测试代码。

import pandas as pd
import numpy as np

#  首先自己定义一些做测试的数据和表头
company, salary, address, experience, education, number_people = '北京知帆科技有限公司', '10.0k-18.0k', '北京-海淀区', '3-4年经验', '本科', '招3人'
data_list = (company, salary, address, experience, education, number_people)
#  tuple   list 类型都可以  区别：一个可变 一个不可变
# data_tuple = [company, salary, address, experience, education, number_people]
head = ('company', 'salary', 'address', 'experience', 'education', 'number_people')

"""
方法一   传入元组或者列表  
"""
# data = np.array([
#     [1, 2, 3, 4, 6, 6, 7],
#     [1, 2, 3, 4, 6, 6, 7]])
#  data  传入list  和 array类型都可以

df = pd.DataFrame(columns=head, data=list([data_list]))
# df = pd.DataFrame(columns=head, data=data)
df.to_csv('aa11.csv', mode='w', index=False, sep=',')

"""
方法二   传入字典
"""
# dic1 = {'company': ['北京知帆科技有限公司'], 'salary': ['10.0k-18.0k'], 'address': ['北京-海淀区'],
#         'experience': ['3-4年经验'], 'education': ['本科'], 'number_people': ['招3人']}
#  字典打包
dic = dict(zip(['company', 'salary', 'address', 'experience', 'education', 'number_people'],
               [[company], [salary], [address], [experience], [education], [number_people]]))

df = pd.DataFrame(columns=head, data=dic)
# df = pd.DataFrame(columns=head, data=dic1)
df.to_csv('aa22.csv', mode='w', header=True, index=False, sep=',')

"""
方法三  文件读写
"""
import codecs
import csv

# 文件读尽量用codecs.open方法，一般不会出现编码的问题。
f = codecs.open('aa33.csv', 'w', encoding='utf-8')
writer = csv.writer(f)
writer.writerow(head)  # 写入表头  也就是文件标题
data_list = ['北京知帆科技有限公司', '10.0k-18.0k', '北京-海淀区', '3-4年经验', '本科', '招3人']
writer.writerow(data_list)

#  如果codes.open用不习惯的话    直接用with open
data_list = ['北京知帆科技有限公司', '10.0k-18.0k', '北京-海淀区', '3-4年经验', '本科', '招3人']
with open('aa44.csv', 'w', encoding='utf-8', newline='') as file:
    writer = csv.writer(file)
    writer.writerow(head)  # 写入表头  也就是文件标题
    writer.writerow(data_list)

下面分享一下爬虫+写入csv的代码
爬虫+pandas写入csv，参考另一篇文章中的代码
爬虫+pandas写入csv

import requests
from bs4 import BeautifulSoup
import json
import csv


def get_city_aqi(pinyin):
    url = 'http://www.pm25.in/' + pinyin
    r = requests.get(url, timeout=60)
    soup = BeautifulSoup(r.text, 'lxml')
    div_list = soup.find_all('div', {'class': 'span1'})

    city_aqi = []
    for i in range(8):
        div_content = div_list[i]
        caption = div_content.find('div', {'class': 'caption'}).text.strip()
        value = div_content.find('div', {'class': 'value'}).text.strip()
        # city_aqi.append((caption, value))
        city_aqi.append(value)
    return city_aqi


def get_all_cities():
    url = 'http://www.pm25.in/'
    city_list = []
    r = requests.get(url, timeout=60)
    soup = BeautifulSoup(r.text, 'lxml')
    city_div = soup.find_all('div', {'class': 'bottom'})[1]
    city_link_list = city_div.find_all('a')
    for city_link in city_link_list:
        city_name = city_link.text
        city_pinyin = city_link['href'][1:]
        city_list.append((city_name, city_pinyin))
    return city_list


def main():
    city_list = get_all_cities()
    header = ['City', 'AQI', 'PM2.5/h', 'PM10/h', 'CO/h', 'NO2/h', 'O3/h', 'O3/8h', 'SO2/h']
    with open('./china_city_aqi.csv', 'w', encoding='utf-8', newline='')as f:
        writer = csv.writer(f)
        writer.writerow(header)
        for i, city in enumerate(city_list):
            print(f'处理第{i + 1}条, 共{len(city_list)}条')
            city_name = city[0]
            city_pinyin = city[1]
            city_aqi = get_city_aqi(city_pinyin)
            row = [city_name] + city_aqi
            print(row)
            writer.writerow(row)


if __name__ == '__main__':
    main()

LSTM时间序列预测pm2.5

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
from datetime import datetime
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.utils import shuffle

data = pd.read_csv('./dataset/data.csv')

data
Noyearmonthdayhourpm2.5DEWPTEMPPREScbwdIwsIsIr
012015110NaN-21-11.01021.0NW1.7900
122015111NaN-21-12.01020.0NW4.9200
232015112NaN-21-11.01019.0NW6.7100
342015113NaN-21-14.01019.0NW9.8400
452015114NaN-20-12.01018.0NW12.9700
..........................................
438194382020191231198.0-23-2.01034.0NW231.9700
4382043821201912312010.0-22-3.01034.0NW237.7800
4382143822201912312110.0-22-3.01034.0NW242.7000
438224382320191231228.0-22-4.01034.0NW246.7200
4382343824201912312312.0-21-3.01034.0NW249.8500

43824 rows × 13 columns

data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 43824 entries, 0 to 43823
Data columns (total 13 columns):
No       43824 non-null int64
year     43824 non-null int64
month    43824 non-null int64
day      43824 non-null int64
hour     43824 non-null int64
pm2.5    41757 non-null float64
DEWP     43824 non-null int64
TEMP     43824 non-null float64
PRES     43824 non-null float64
cbwd     43824 non-null object
Iws      43824 non-null float64
Is       43824 non-null int64
Ir       43824 non-null int64
dtypes: float64(4), int64(8), object(1)
memory usage: 4.3+ MB

data['pm2.5'].isna().sum()

2067

数据处理

pm2.5有空缺 前面的删除 后面的填充 保证是一个完整的序列

data = data.iloc[24:, :].fillna(method='ffill')  # 使用前向填充 空值的前一个值填充
data
Noyearmonthdayhourpm2.5DEWPTEMPPREScbwdIwsIsIr
24252015120129.0-16-4.01020.0SE1.7900
25262015121148.0-15-4.01020.0SE2.6800
26272015122159.0-11-5.01021.0SE3.5700
27282015123181.0-7-5.01022.0SE5.3610
28292015124138.0-7-5.01022.0SE6.2520
..........................................
438194382020191231198.0-23-2.01034.0NW231.9700
4382043821201912312010.0-22-3.01034.0NW237.7800
4382143822201912312110.0-22-3.01034.0NW242.7000
438224382320191231228.0-22-4.01034.0NW246.7200
4382343824201912312312.0-21-3.01034.0NW249.8500

43800 rows × 13 columns

plt.figure(figsize=(12, 5))
# 查看最后1000次pm2.5观测状态
plt.subplot(1, 2, 1)
data['pm2.5'][-1000:].plot()
plt.title('pm2.5')
# 查看最后1000次温度观测状态
plt.subplot(1, 2, 2)
plt.title('TEMP')
data['TEMP'][-1000:].plot()
plt.show()

在这里插入图片描述

时间处理

对时间做处理，将时间合并作为index

data['time'] = data.apply(lambda x: datetime(x['year'], x['month'], x['day'], x['hour']), axis=1)
data.drop(['year', 'month', 'day', 'hour', 'No'], axis=1, inplace=True)
data.set_index('time', inplace=True)
data
pm2.5DEWPTEMPPREScbwdIwsIsIr
time
2015-01-02 00:00:00129.0-16-4.01020.0SE1.7900
2015-01-02 01:00:00148.0-15-4.01020.0SE2.6800
2015-01-02 02:00:00159.0-11-5.01021.0SE3.5700
2015-01-02 03:00:00181.0-7-5.01022.0SE5.3610
2015-01-02 04:00:00138.0-7-5.01022.0SE6.2520
...........................
2019-12-31 19:00:008.0-23-2.01034.0NW231.9700
2019-12-31 20:00:0010.0-22-3.01034.0NW237.7800
2019-12-31 21:00:0010.0-22-3.01034.0NW242.7000
2019-12-31 22:00:008.0-22-4.01034.0NW246.7200
2019-12-31 23:00:0012.0-21-3.01034.0NW249.8500

43800 rows × 8 columns

# 将 cbwd 风向one_hot编码
data = data.join(pd.get_dummies(data['cbwd']))
del data['cbwd']
data
pm2.5DEWPTEMPPRESIwsIsIrNENWSEcv
time
2015-01-02 00:00:00129.0-16-4.01020.01.79000010
2015-01-02 01:00:00148.0-15-4.01020.02.68000010
2015-01-02 02:00:00159.0-11-5.01021.03.57000010
2015-01-02 03:00:00181.0-7-5.01022.05.36100010
2015-01-02 04:00:00138.0-7-5.01022.06.25200010
....................................
2019-12-31 19:00:008.0-23-2.01034.0231.97000100
2019-12-31 20:00:0010.0-22-3.01034.0237.78000100
2019-12-31 21:00:0010.0-22-3.01034.0242.70000100
2019-12-31 22:00:008.0-22-4.01034.0246.72000100
2019-12-31 23:00:0012.0-21-3.01034.0249.85000100

43800 rows × 11 columns

plt.figure(figsize=(12, 5))
# 查看最后1000次pm2.5观测状态
plt.subplot(1, 2, 1)
data['pm2.5'][-1000:].plot()
plt.title('pm2.5')
# 查看最后1000次温度观测状态
plt.subplot(1, 2, 2)
plt.title('TEMP')
data['TEMP'][-1000:].plot()
plt.show()

在这里插入图片描述

# 从序列中提取train test
seq_length = 5 * 24  # 提前每个时刻前五天的数据
delay = 24  # 预测第六天的数据

# 先将 每隔前6天的数据提取出来
_data_list = []
for i in range(len(data)-seq_length-delay):
    _data_list.append(data.iloc[i: i+seq_length+delay])
print(_data_list[0].shape) 

(144, 11)

# 将dataframe转成array
_data = np.array([df.values for df in _data_list])
_data.shape

(43656, 144, 11)

# 划分训练集和测试集
x = _data[:, :seq_length, 1:]
y = _data[:, -1, 0]

X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=1000)
print(X_train.shape)  
print(y_train.shape)  

(34924, 120, 10)
(34924,)

# 数据标准化
mean = X_train.mean(axis=0)
std = X_train.std(axis=0)
X_train = (X_train - mean)/std
X_test = (X_test - mean)/std

from tensorflow import keras
batch_size = 128
# 使用多层lstm (该数据有时间变化趋势, 适合使用lstm)
model = keras.Sequential()
# model.add(keras.layers.LSTM(32, input_shape=[X_train[1:]], return_sequences=True))
model.add(keras.layers.LSTM(32, return_sequences=True))  # 有4个神经元数量为32的前馈网络层
model.add(keras.layers.LSTM(32, return_sequences=True))
model.add(keras.layers.LSTM(32, return_sequences=True))
model.add(keras.layers.LSTM(32, return_sequences=False))
model.add(keras.layers.Dense(1))
# 优化  训练过程中降低学习速率
# 理解： 连续3个epoch学习速率没下降 lr*0.5  最后不超过最低值
lr_reduce = keras.callbacks.ReduceLROnPlateau(monitor='val_loss', patience=3, factor=0.5,
                                  min_lr=0.00001)
model.compile(optimizer='adam', loss='mse', metrics=['mae'])  # mae平均绝对误差
history = model.fit(X_train, y_train, batch_size=batch_size,
                    epochs=150,
                    callbacks=[lr_reduce],
                    validation_data=(X_test, y_test))
# model.save('./pm2.5.h5')
print(history.history.keys())
plt.plot(history.epoch, history.history['mae'], c='r')
plt.plot(history.epoch, history.history['val_mae'], c='g')
plt.legend()
plt.show()
plt.plot(history.epoch, history.history['loss'], c='r', label='Training loss')
plt.plot(history.epoch, history.history['val_loss'], c='g', label='Validation loss')
plt.title('Training and validation loss')
plt.legend()
plt.show()

Epoch 1/150
273/273 [==============================] - 61s 206ms/step - loss: 16742.4334 - mae: 90.9238 - val_loss: 14646.6338 - val_mae: 81.6788
Epoch 2/150
273/273 [==============================] - 57s 209ms/step - loss: 14607.2083 - mae: 81.2521 - val_loss: 13331.6240 - val_mae: 76.2547
Epoch 3/150
273/273 [==============================] - 58s 212ms/step - loss: 13062.3841 - mae: 75.8968 - val_loss: 12259.8438 - val_mae: 72.3604
Epoch 4/150
273/273 [==============================] - 60s 218ms/step - loss: 12766.5135 - mae: 73.8100 - val_loss: 11373.2275 - val_mae: 69.4936
Epoch 5/150
273/273 [==============================] - 63s 230ms/step - loss: 11479.0624 - mae: 70.2214 - val_loss: 10639.9043 - val_mae: 67.4292
Epoch 6/150
273/273 [==============================] - 62s 228ms/step - loss: 10776.3757 - mae: 68.3468 - val_loss: 10039.1123 - val_mae: 65.9866
Epoch 7/150
273/273 [==============================] - 62s 227ms/step - loss: 10341.7646 - mae: 67.2075 - val_loss: 9554.6572 - val_mae: 65.0422
Epoch 8/150
273/273 [==============================] - 61s 224ms/step - loss: 9624.0328 - mae: 65.5107 - val_loss: 9170.4385 - val_mae: 64.5219
Epoch 9/150
273/273 [==============================] - 60s 221ms/step - loss: 9403.6663 - mae: 65.6162 - val_loss: 8875.7959 - val_mae: 64.3700
Epoch 10/150
273/273 [==============================] - 61s 222ms/step - loss: 9010.0623 - mae: 65.4931 - val_loss: 8653.1328 - val_mae: 64.4761
Epoch 11/150
273/273 [==============================] - 61s 224ms/step - loss: 8827.4738 - mae: 65.5057 - val_loss: 8494.2217 - val_mae: 64.7843
Epoch 12/150
273/273 [==============================] - 60s 220ms/step - loss: 8426.1633 - mae: 65.4281 - val_loss: 8383.6738 - val_mae: 65.2093
Epoch 13/150
273/273 [==============================] - 60s 220ms/step - loss: 8401.8991 - mae: 65.6818 - val_loss: 8311.4551 - val_mae: 65.6845
Epoch 14/150
273/273 [==============================] - 60s 220ms/step - loss: 8488.3962 - mae: 66.8363 - val_loss: 8268.0596 - val_mae: 66.1579
Epoch 15/150
273/273 [==============================] - 60s 220ms/step - loss: 8513.0347 - mae: 67.4366 - val_loss: 8243.9941 - val_mae: 66.5921
Epoch 16/150
273/273 [==============================] - 60s 222ms/step - loss: 8277.4878 - mae: 67.0897 - val_loss: 8232.0371 - val_mae: 66.9601
Epoch 17/150
273/273 [==============================] - 60s 220ms/step - loss: 8663.8273 - mae: 68.2389 - val_loss: 8226.9561 - val_mae: 67.2551
Epoch 18/150
273/273 [==============================] - 61s 222ms/step - loss: 8288.9277 - mae: 67.6184 - val_loss: 8225.4375 - val_mae: 67.4816
Epoch 19/150
273/273 [==============================] - 61s 224ms/step - loss: 8567.5850 - mae: 68.9895 - val_loss: 8225.4082 - val_mae: 67.6185
Epoch 20/150
273/273 [==============================] - 61s 222ms/step - loss: 8487.0704 - mae: 68.8503 - val_loss: 8225.7344 - val_mae: 67.7098
Epoch 21/150
273/273 [==============================] - 61s 223ms/step - loss: 8613.7959 - mae: 69.0857 - val_loss: 8226.1621 - val_mae: 67.7786
Epoch 22/150
273/273 [==============================] - 60s 220ms/step - loss: 8356.8230 - mae: 68.3529 - val_loss: 8226.2686 - val_mae: 67.7927
Epoch 23/150
273/273 [==============================] - 60s 219ms/step - loss: 8468.6951 - mae: 68.4023 - val_loss: 8226.3145 - val_mae: 67.7986
Epoch 24/150
273/273 [==============================] - 60s 221ms/step - loss: 8291.6117 - mae: 68.0808 - val_loss: 8226.4033 - val_mae: 67.8098
Epoch 25/150
273/273 [==============================] - 61s 223ms/step - loss: 8332.3899 - mae: 68.5825 - val_loss: 8226.5439 - val_mae: 67.8266
Epoch 26/150
273/273 [==============================] - 61s 222ms/step - loss: 8465.6008 - mae: 68.7638 - val_loss: 8226.5449 - val_mae: 67.8265
Epoch 27/150
273/273 [==============================] - 61s 225ms/step - loss: 8370.8415 - mae: 68.7813 - val_loss: 8226.5410 - val_mae: 67.8264
Epoch 28/150
273/273 [==============================] - 61s 225ms/step - loss: 8463.0080 - mae: 68.5800 - val_loss: 8226.5957 - val_mae: 67.8325
Epoch 29/150
273/273 [==============================] - 63s 231ms/step - loss: 8455.1988 - mae: 68.3467 - val_loss: 8226.5693 - val_mae: 67.8294
Epoch 30/150
273/273 [==============================] - 61s 222ms/step - loss: 8236.1620 - mae: 67.9002 - val_loss: 8226.5703 - val_mae: 67.8297
Epoch 31/150
273/273 [==============================] - 60s 219ms/step - loss: 8181.2296 - mae: 67.7665 - val_loss: 8226.5859 - val_mae: 67.8314
Epoch 32/150
273/273 [==============================] - 60s 219ms/step - loss: 8321.2633 - mae: 68.2151 - val_loss: 8226.5898 - val_mae: 67.8316
Epoch 33/150
273/273 [==============================] - 61s 222ms/step - loss: 8462.1091 - mae: 68.9432 - val_loss: 8226.6045 - val_mae: 67.8336
Epoch 34/150
273/273 [==============================] - 61s 224ms/step - loss: 8383.3842 - mae: 68.7199 - val_loss: 8226.5908 - val_mae: 67.8318
Epoch 35/150
273/273 [==============================] - 61s 224ms/step - loss: 8547.4360 - mae: 68.6443 - val_loss: 8226.5996 - val_mae: 67.8329
Epoch 36/150
273/273 [==============================] - 62s 227ms/step - loss: 8406.0428 - mae: 68.6697 - val_loss: 8226.5967 - val_mae: 67.8328
Epoch 37/150
273/273 [==============================] - 61s 225ms/step - loss: 8370.8651 - mae: 68.4359 - val_loss: 8226.5947 - val_mae: 67.8324
Epoch 38/150
273/273 [==============================] - 62s 226ms/step - loss: 8351.3386 - mae: 68.5521 - val_loss: 8226.5957 - val_mae: 67.8324
Epoch 39/150
273/273 [==============================] - 61s 222ms/step - loss: 8484.4870 - mae: 68.4776 - val_loss: 8226.5947 - val_mae: 67.8323
Epoch 40/150
273/273 [==============================] - 61s 222ms/step - loss: 8532.6639 - mae: 68.9514 - val_loss: 8226.5986 - val_mae: 67.8328
Epoch 41/150
273/273 [==============================] - 61s 225ms/step - loss: 8536.1996 - mae: 68.9368 - val_loss: 8226.5947 - val_mae: 67.8323
Epoch 42/150
273/273 [==============================] - 61s 222ms/step - loss: 8424.5842 - mae: 68.4884 - val_loss: 8226.5967 - val_mae: 67.8325
Epoch 43/150
273/273 [==============================] - 60s 221ms/step - loss: 8510.1979 - mae: 68.9535 - val_loss: 8226.5967 - val_mae: 67.8326
Epoch 44/150
273/273 [==============================] - 60s 220ms/step - loss: 8414.9673 - mae: 68.4775 - val_loss: 8226.5967 - val_mae: 67.8326
Epoch 45/150
273/273 [==============================] - 60s 221ms/step - loss: 8424.7148 - mae: 68.5667 - val_loss: 8226.5986 - val_mae: 67.8328
Epoch 46/150
273/273 [==============================] - 63s 231ms/step - loss: 8584.4803 - mae: 69.1215 - val_loss: 8226.5977 - val_mae: 67.8326
Epoch 47/150
273/273 [==============================] - 61s 224ms/step - loss: 8676.9013 - mae: 69.2567 - val_loss: 8226.5977 - val_mae: 67.8328
Epoch 48/150
273/273 [==============================] - 61s 224ms/step - loss: 8390.5315 - mae: 68.5066 - val_loss: 8226.5947 - val_mae: 67.8324
Epoch 49/150
273/273 [==============================] - 61s 224ms/step - loss: 8349.9124 - mae: 68.4460 - val_loss: 8226.5986 - val_mae: 67.8327
Epoch 50/150
273/273 [==============================] - 60s 221ms/step - loss: 8205.4177 - mae: 67.7604 - val_loss: 8226.5967 - val_mae: 67.8325
Epoch 51/150
273/273 [==============================] - 61s 223ms/step - loss: 8392.1621 - mae: 68.7458 - val_loss: 8226.5957 - val_mae: 67.8326
Epoch 52/150
273/273 [==============================] - 61s 222ms/step - loss: 8675.7586 - mae: 69.4464 - val_loss: 8226.5967 - val_mae: 67.8326
Epoch 53/150
273/273 [==============================] - 62s 226ms/step - loss: 8376.2547 - mae: 68.5158 - val_loss: 8226.5986 - val_mae: 67.8327
Epoch 54/150
273/273 [==============================] - 62s 227ms/step - loss: 8335.3379 - mae: 68.3225 - val_loss: 8226.5986 - val_mae: 67.8327
Epoch 55/150
273/273 [==============================] - 62s 226ms/step - loss: 8481.7657 - mae: 68.7639 - val_loss: 8226.5986 - val_mae: 67.8328
Epoch 56/150
273/273 [==============================] - 63s 230ms/step - loss: 8512.7698 - mae: 68.5890 - val_loss: 8226.5957 - val_mae: 67.8325
Epoch 57/150
273/273 [==============================] - 63s 230ms/step - loss: 8455.3750 - mae: 68.6572 - val_loss: 8226.5977 - val_mae: 67.8328
Epoch 58/150
273/273 [==============================] - 61s 223ms/step - loss: 8470.7624 - mae: 68.9075 - val_loss: 8226.5986 - val_mae: 67.8327
Epoch 59/150
273/273 [==============================] - 60s 221ms/step - loss: 8479.7735 - mae: 68.9081 - val_loss: 8226.5947 - val_mae: 67.8323
Epoch 60/150
273/273 [==============================] - 61s 224ms/step - loss: 8234.5817 - mae: 68.7015 - val_loss: 8226.6016 - val_mae: 67.8330
Epoch 61/150
273/273 [==============================] - 60s 222ms/step - loss: 8461.2091 - mae: 68.7669 - val_loss: 8226.5986 - val_mae: 67.8327
Epoch 62/150
273/273 [==============================] - 60s 222ms/step - loss: 8458.8530 - mae: 68.7402 - val_loss: 8226.5977 - val_mae: 67.8327
Epoch 63/150
273/273 [==============================] - 62s 226ms/step - loss: 8587.8940 - mae: 68.6250 - val_loss: 8226.6006 - val_mae: 67.8331
Epoch 64/150
273/273 [==============================] - 62s 226ms/step - loss: 8602.9138 - mae: 68.8073 - val_loss: 8226.5957 - val_mae: 67.8328
Epoch 65/150
273/273 [==============================] - 61s 224ms/step - loss: 8389.7711 - mae: 68.1815 - val_loss: 8226.5078 - val_mae: 67.8320
Epoch 66/150
273/273 [==============================] - 62s 225ms/step - loss: 8585.8489 - mae: 69.0446 - val_loss: 8225.7852 - val_mae: 67.8145
Epoch 67/150
273/273 [==============================] - 62s 228ms/step - loss: 8323.5853 - mae: 68.4999 - val_loss: 8225.4238 - val_mae: 67.8042
Epoch 68/150
273/273 [==============================] - 62s 228ms/step - loss: 8545.9198 - mae: 68.8307 - val_loss: 8228.8105 - val_mae: 67.8080
Epoch 69/150
273/273 [==============================] - 62s 226ms/step - loss: 8262.9887 - mae: 67.9555 - val_loss: 8223.8291 - val_mae: 67.7737
Epoch 70/150
273/273 [==============================] - 61s 224ms/step - loss: 8656.3001 - mae: 69.2458 - val_loss: 8222.8906 - val_mae: 67.7547
Epoch 71/150
273/273 [==============================] - 61s 224ms/step - loss: 8336.8825 - mae: 68.4149 - val_loss: 8220.6260 - val_mae: 67.7138
Epoch 72/150
273/273 [==============================] - 61s 224ms/step - loss: 8477.9109 - mae: 68.8455 - val_loss: 8218.2393 - val_mae: 67.6565
Epoch 73/150
273/273 [==============================] - 62s 226ms/step - loss: 8585.9266 - mae: 68.7750 - val_loss: 8214.5684 - val_mae: 67.5979
Epoch 74/150
273/273 [==============================] - 61s 224ms/step - loss: 8216.9939 - mae: 67.7797 - val_loss: 8208.9502 - val_mae: 67.4845
Epoch 75/150
273/273 [==============================] - 61s 224ms/step - loss: 8393.6422 - mae: 68.2826 - val_loss: 8196.3428 - val_mae: 67.3194
Epoch 76/150
273/273 [==============================] - 61s 225ms/step - loss: 8130.0958 - mae: 67.4432 - val_loss: 8181.2900 - val_mae: 67.1320
Epoch 77/150
273/273 [==============================] - 82s 300ms/step - loss: 8471.2442 - mae: 67.8909 - val_loss: 8160.0679 - val_mae: 66.8343
Epoch 78/150
273/273 [==============================] - 61s 222ms/step - loss: 8297.1601 - mae: 67.2958 - val_loss: 8129.6997 - val_mae: 66.4568
Epoch 79/150
273/273 [==============================] - 60s 221ms/step - loss: 8331.4552 - mae: 67.1481 - val_loss: 8101.7114 - val_mae: 66.1418
Epoch 80/150
273/273 [==============================] - 61s 222ms/step - loss: 8084.1070 - mae: 66.4432 - val_loss: 8080.2290 - val_mae: 66.0347
Epoch 81/150
273/273 [==============================] - 61s 223ms/step - loss: 8396.0239 - mae: 67.0763 - val_loss: 8069.1743 - val_mae: 65.8770
Epoch 82/150
273/273 [==============================] - 60s 219ms/step - loss: 8185.0534 - mae: 66.4298 - val_loss: 8060.0220 - val_mae: 65.9004
Epoch 83/150
273/273 [==============================] - 60s 220ms/step - loss: 8303.1067 - mae: 66.9463 - val_loss: 8052.3354 - val_mae: 65.8416
Epoch 84/150
273/273 [==============================] - 61s 223ms/step - loss: 8047.7834 - mae: 65.8863 - val_loss: 8045.4146 - val_mae: 65.8027
Epoch 85/150
273/273 [==============================] - 61s 222ms/step - loss: 8329.9191 - mae: 66.8130 - val_loss: 8035.4189 - val_mae: 65.6562
Epoch 86/150
273/273 [==============================] - 60s 220ms/step - loss: 8186.9528 - mae: 66.5324 - val_loss: 8027.5225 - val_mae: 65.5805
Epoch 87/150
273/273 [==============================] - 60s 221ms/step - loss: 8057.1022 - mae: 65.9535 - val_loss: 8019.4668 - val_mae: 65.5211
Epoch 88/150
273/273 [==============================] - 61s 223ms/step - loss: 8341.6970 - mae: 66.4929 - val_loss: 8012.3262 - val_mae: 65.4549
Epoch 89/150
273/273 [==============================] - 61s 225ms/step - loss: 8027.3912 - mae: 65.7953 - val_loss: 8005.6152 - val_mae: 65.4042
Epoch 90/150
273/273 [==============================] - 61s 225ms/step - loss: 8064.1161 - mae: 66.0203 - val_loss: 7999.6040 - val_mae: 65.2895
Epoch 91/150
273/273 [==============================] - 62s 226ms/step - loss: 8208.3036 - mae: 66.3922 - val_loss: 7995.8965 - val_mae: 65.3147
Epoch 92/150
273/273 [==============================] - 60s 221ms/step - loss: 8032.6061 - mae: 65.7072 - val_loss: 7989.3223 - val_mae: 65.2171
Epoch 93/150
273/273 [==============================] - 61s 222ms/step - loss: 7989.7291 - mae: 65.4608 - val_loss: 7983.4604 - val_mae: 65.2041
Epoch 94/150
273/273 [==============================] - 61s 224ms/step - loss: 8133.9053 - mae: 66.1046 - val_loss: 7980.9595 - val_mae: 65.1901
Epoch 95/150
273/273 [==============================] - 62s 226ms/step - loss: 7963.0258 - mae: 65.6962 - val_loss: 7975.8901 - val_mae: 65.1806
Epoch 96/150
273/273 [==============================] - 62s 226ms/step - loss: 8335.9137 - mae: 66.2708 - val_loss: 7970.1880 - val_mae: 65.0943
Epoch 97/150
273/273 [==============================] - 79s 290ms/step - loss: 8136.6836 - mae: 66.1052 - val_loss: 7964.5376 - val_mae: 65.0385
Epoch 98/150
273/273 [==============================] - 61s 224ms/step - loss: 8103.1769 - mae: 65.9585 - val_loss: 7962.0000 - val_mae: 65.0376
Epoch 99/150
273/273 [==============================] - 60s 218ms/step - loss: 8051.2684 - mae: 65.7560 - val_loss: 7956.1284 - val_mae: 64.9462
Epoch 100/150
273/273 [==============================] - 61s 222ms/step - loss: 8102.4803 - mae: 65.4313 - val_loss: 7951.5518 - val_mae: 64.9178
Epoch 101/150
273/273 [==============================] - 61s 222ms/step - loss: 8274.3461 - mae: 65.8671 - val_loss: 7946.9150 - val_mae: 64.8726
Epoch 102/150
273/273 [==============================] - 61s 223ms/step - loss: 8052.9716 - mae: 65.5578 - val_loss: 7943.2925 - val_mae: 64.8318
Epoch 103/150
273/273 [==============================] - 61s 224ms/step - loss: 8165.8697 - mae: 65.6790 - val_loss: 7936.5962 - val_mae: 64.7525
Epoch 104/150
273/273 [==============================] - 61s 223ms/step - loss: 8129.4666 - mae: 65.5902 - val_loss: 7931.2388 - val_mae: 64.6802
Epoch 105/150
273/273 [==============================] - 61s 222ms/step - loss: 8077.7160 - mae: 65.1185 - val_loss: 7925.7744 - val_mae: 64.6451
Epoch 106/150
273/273 [==============================] - 66s 241ms/step - loss: 8382.6036 - mae: 66.0239 - val_loss: 7919.8882 - val_mae: 64.5722
Epoch 107/150
273/273 [==============================] - 70s 256ms/step - loss: 8135.8677 - mae: 65.2656 - val_loss: 7916.8335 - val_mae: 64.5181
Epoch 108/150
273/273 [==============================] - 71s 258ms/step - loss: 8049.9404 - mae: 65.0456 - val_loss: 7908.3511 - val_mae: 64.4399
Epoch 109/150
273/273 [==============================] - 72s 264ms/step - loss: 7939.6620 - mae: 65.1306 - val_loss: 7901.5796 - val_mae: 64.3617
Epoch 110/150
273/273 [==============================] - 70s 255ms/step - loss: 8059.7353 - mae: 65.2351 - val_loss: 7894.9629 - val_mae: 64.2769
Epoch 111/150
273/273 [==============================] - 71s 259ms/step - loss: 8078.8482 - mae: 64.9532 - val_loss: 7889.0737 - val_mae: 64.1602
Epoch 112/150
273/273 [==============================] - 71s 259ms/step - loss: 8203.2952 - mae: 65.1534 - val_loss: 7881.8872 - val_mae: 64.1428
Epoch 113/150
273/273 [==============================] - 71s 260ms/step - loss: 8021.4052 - mae: 64.7163 - val_loss: 7875.6646 - val_mae: 64.0615
Epoch 114/150
273/273 [==============================] - 71s 258ms/step - loss: 8201.4341 - mae: 65.2901 - val_loss: 7867.0938 - val_mae: 63.9763
Epoch 115/150
273/273 [==============================] - 72s 263ms/step - loss: 8111.1570 - mae: 64.7002 - val_loss: 7860.6860 - val_mae: 63.8970
Epoch 116/150
273/273 [==============================] - 69s 253ms/step - loss: 7792.6684 - mae: 64.3929 - val_loss: 7853.4888 - val_mae: 63.8189
Epoch 117/150
273/273 [==============================] - 68s 250ms/step - loss: 8090.4443 - mae: 64.7919 - val_loss: 7846.6753 - val_mae: 63.7234
Epoch 118/150
273/273 [==============================] - 68s 250ms/step - loss: 8018.4777 - mae: 64.5421 - val_loss: 7838.5083 - val_mae: 63.6637
Epoch 119/150
273/273 [==============================] - 70s 255ms/step - loss: 8078.1974 - mae: 64.7491 - val_loss: 7832.1821 - val_mae: 63.6111
Epoch 120/150
273/273 [==============================] - 69s 252ms/step - loss: 7839.0215 - mae: 63.8854 - val_loss: 7820.8467 - val_mae: 63.5104
Epoch 121/150
273/273 [==============================] - 68s 249ms/step - loss: 7987.1598 - mae: 64.1657 - val_loss: 7815.5684 - val_mae: 63.5297
Epoch 122/150
273/273 [==============================] - 68s 248ms/step - loss: 8036.7563 - mae: 64.6019 - val_loss: 7803.7461 - val_mae: 63.3131
Epoch 123/150
273/273 [==============================] - 68s 250ms/step - loss: 7949.9265 - mae: 63.8457 - val_loss: 7795.9517 - val_mae: 63.2526
Epoch 124/150
273/273 [==============================] - 71s 261ms/step - loss: 8143.0449 - mae: 64.3155 - val_loss: 7789.0024 - val_mae: 63.1432
Epoch 125/150
273/273 [==============================] - 68s 250ms/step - loss: 8140.8686 - mae: 64.2501 - val_loss: 7782.2739 - val_mae: 63.1526
Epoch 126/150
273/273 [==============================] - 69s 253ms/step - loss: 8049.5511 - mae: 64.2005 - val_loss: 7775.1011 - val_mae: 63.0455
Epoch 127/150
273/273 [==============================] - 69s 251ms/step - loss: 7764.5967 - mae: 63.6608 - val_loss: 7768.4590 - val_mae: 63.0593
Epoch 128/150
273/273 [==============================] - 67s 245ms/step - loss: 7836.4137 - mae: 63.7229 - val_loss: 7761.5576 - val_mae: 62.9700
Epoch 129/150
273/273 [==============================] - 67s 246ms/step - loss: 7844.0585 - mae: 63.6464 - val_loss: 7756.0679 - val_mae: 63.0015
Epoch 130/150
273/273 [==============================] - 66s 244ms/step - loss: 7920.9684 - mae: 64.0747 - val_loss: 7753.7666 - val_mae: 62.9503
Epoch 131/150
273/273 [==============================] - 68s 248ms/step - loss: 7833.7799 - mae: 63.6380 - val_loss: 7747.2690 - val_mae: 62.8746
Epoch 132/150
273/273 [==============================] - 68s 248ms/step - loss: 7904.2761 - mae: 63.7532 - val_loss: 7742.7090 - val_mae: 62.8583
Epoch 133/150
273/273 [==============================] - 69s 254ms/step - loss: 8047.6140 - mae: 63.9913 - val_loss: 7737.4053 - val_mae: 62.7785
Epoch 134/150
273/273 [==============================] - 67s 246ms/step - loss: 7943.3913 - mae: 63.6909 - val_loss: 7732.6128 - val_mae: 62.7872
Epoch 135/150
273/273 [==============================] - 68s 248ms/step - loss: 7937.4888 - mae: 63.6617 - val_loss: 7728.6724 - val_mae: 62.6928
Epoch 136/150
273/273 [==============================] - 62s 228ms/step - loss: 7955.2694 - mae: 63.7793 - val_loss: 7724.0063 - val_mae: 62.6431
Epoch 137/150
273/273 [==============================] - 62s 226ms/step - loss: 8021.3685 - mae: 63.9636 - val_loss: 7719.7700 - val_mae: 62.6444
Epoch 138/150
273/273 [==============================] - 61s 225ms/step - loss: 7850.8130 - mae: 63.3130 - val_loss: 7714.8394 - val_mae: 62.6192
Epoch 139/150
273/273 [==============================] - 61s 225ms/step - loss: 7899.5321 - mae: 63.5370 - val_loss: 7710.7456 - val_mae: 62.5520
Epoch 140/150
273/273 [==============================] - 61s 222ms/step - loss: 8138.2508 - mae: 64.0794 - val_loss: 7711.3955 - val_mae: 62.5012
Epoch 141/150
273/273 [==============================] - 60s 222ms/step - loss: 7783.9759 - mae: 63.3851 - val_loss: 7705.9009 - val_mae: 62.5233
Epoch 142/150
273/273 [==============================] - 61s 223ms/step - loss: 7844.7223 - mae: 63.2917 - val_loss: 7701.9321 - val_mae: 62.5271
Epoch 143/150
273/273 [==============================] - 60s 220ms/step - loss: 7773.0852 - mae: 62.9951 - val_loss: 7697.3213 - val_mae: 62.4414
Epoch 144/150
273/273 [==============================] - 61s 222ms/step - loss: 7961.3421 - mae: 63.7030 - val_loss: 7691.9458 - val_mae: 62.3815
Epoch 145/150
273/273 [==============================] - 60s 222ms/step - loss: 7652.2020 - mae: 62.5647 - val_loss: 7687.1699 - val_mae: 62.3395
Epoch 146/150
273/273 [==============================] - 61s 224ms/step - loss: 7828.0561 - mae: 63.0869 - val_loss: 7683.5947 - val_mae: 62.3607
Epoch 147/150
273/273 [==============================] - 60s 220ms/step - loss: 7741.2522 - mae: 62.8325 - val_loss: 7679.0864 - val_mae: 62.2841
Epoch 148/150
273/273 [==============================] - 60s 222ms/step - loss: 7799.2858 - mae: 62.9819 - val_loss: 7680.0063 - val_mae: 62.3857
Epoch 149/150
273/273 [==============================] - 60s 222ms/step - loss: 7844.7404 - mae: 63.3468 - val_loss: 7672.7368 - val_mae: 62.2881
Epoch 150/150
273/273 [==============================] - 61s 224ms/step - loss: 7807.6913 - mae: 62.7041 - val_loss: 7667.2842 - val_mae: 62.1901


No handles with labels found to put in legend.


dict_keys(['loss', 'mae', 'val_loss', 'val_mae', 'lr'])

在这里插入图片描述

# 预测1
predict = model.predict(X_test[:5])
print('预测结果为：', predict)
print('真实结果为：', y_test[:5])

预测结果为： [[ 98.9861  ]
 [ 55.642155]
 [102.504036]
 [ 58.44735 ]
 [101.403114]]
真实结果为： [ 86.  89. 151.  41. 121.]

# 预测2 用前五天数据预测第六天数据
data_test = pd.read_csv('./dataset/data.csv')
data_test = data_test.iloc[-120:]
# print(data_test.head())
data_test = data_test.iloc[:, 6:]
# print(data_test)
data_test = data_test.join(pd.get_dummies(data_test['cbwd']))
del data_test['cbwd']
# print(data_test)
# 归一化  使用训练集的mean  std
data_test = (data_test-mean)/std
# data_test = data_test.to_numpy() 
data_test = np.array(data_test)  
print(data_test.shape)
# 将数据扩展成三维
# data_test_expand = np.expand_dims(data_test, 0) 
data_test_expand = data_test.reshape((1,)+(data_test.shape))
print(data_test_expand.shape)
# load_model = keras.models.load_model('./pm2.5.h5')
# predict1 = load_model.predict(data_test_expand)
predict1 = model.predict(data_test_expand)
print('北京2020-01 PM2.5预测结果为', predict1)

(120, 10)
(1, 120, 10)
北京2020-01 PM2.5预测结果为 [[58.903873]]

真实数据对比
在这里插入图片描述

BRYTLEVSON
关注
  • 8
    点赞
  • 踩
  • 50
    收藏
    觉得还不错? 一键收藏
  • 打赏
    打赏
  • 3
    评论
专栏目录
使用 Python 爬取天气数据并保存至 CSV
流楚丶格念的博客
04-19 898
天气数据是我们生活中非常重要的一部分。它不仅影响着我们的出行计划，还直接影响到农业、交通运输、航空航天等领域的运营和决策。因此，获取准确、及时的天气数据对于许多人来说是至关重要的。然而，要获取天气数据并不总是一件容易的事情。虽然有许多天气预报网站和应用程序可以提供天气信息，但有时我们需要的是历史天气数据，而这些数据并不总是可以轻易获得。因此，我们可以利用网络爬虫技术，通过编写程序从天气网站上获取历史天气数据，并将其保存至本地文件，以备后续分析和使用。
【项目实战】Python实现循环神经网络SimpleRNN、LSTM进行淘宝商品评论情感分析(含爬虫程序)
胖哥真不错的博客
09-07 8418
说明：这是一个机器学习实战项目（附带数据+代码），如需数据+完整代码可以直接到文章最后获取。 1.项目背景 随着信息化社会的发展，互联网成为方便、快捷的信息获取渠道之一。在电子商务和社会网站中，大量非结构化的评论文本作为最直观的用户体验数据被保存下来。如何利用这些文字信息归纳出用户对某一事、物的观点态度成为自然语言(NLP)领域一项重要任务。 文本情感分析又称文本意见挖掘，是自然语言处理领域最优吸引力的研究方向之一。文本情感分析通过计算语言学知识量化处理得到...
（神经网络）i7处理器日常裂开的一天：从网络爬虫到数据预测作图，基于LSTM和RNN的温度序列预测
qq_45804132的博客
11-28 401
目录爬虫部分：爬取数据及分离数据预测部分：训练及建模绘图 代码已在本人设备上完美的跑过了，如果在一些读者的电脑上有问题，大 家可以检查如下东西： 1. 代码是否拼接在一起，因为实在jupyter上写的，因此我这里不需要刻意拼 接大家可以将代码糅合在一起。 2. torch版本和anconda版本的问题，可能有部分函数或者方法不兼容。 3. matplotlib版本问题，在之前的某次版本更新后，pyplot的部分方法 改变了。 4. 文件的读取形式，我的csv文件的存储形式为ANSI，因此使用UTF-8
python爬虫把数据保存到csv、mysql中
冷巷(✘_✘)的博客
12-24 3680
啧，放假几天游戏玩腻了，啥都不想干，突然想起来python这玩意，无聊就来玩玩目标：起点主要是拿到这几个数据分析下网页一个li对应一本小说，打开li看里面的东西首先是排名xpath来一下名字在第二个div里面接着是作者类型最后是章节和更新时间直接上代码导入库pymysql是连接mysqlrequests是请求库parsel是解析库csv是保存到csv文件里面看一下网址，拿一下请求头，然后直接请求接着解析上面的几个数据。
学了那么久Python还什么都做不了，我觉得你该试试这个方法了
龙叔的博客
11-08 1万+
答应我，别再做无用功了
python爬虫——保存数据为.csv文件
热门推荐
weixin_56067972的博客
11-15 2万+
爬虫存储数据为.csv文件
Python中利用LSTM模型进行时间序列预测分析的实现
09-18
主要介绍了Python中利用LSTM模型进行时间序列预测分析的实现，文中通过示例代码介绍的非常详细，对大家的学习或者工作具有一定的参考学习价值，需要的朋友们下面随着小编来一起学习学习吧
15.时间序列预测（LSTM模型）python代码实现
05-08
时间序列预测（LSTM模型）python代码实现"这个主题涵盖了使用Python和LSTM进行时间序列预测的全过程，包括理解时间序列数据、搭建LSTM模型、数据预处理、模型训练、验证和预测。通过实践这一过程，可以提升对LSTM和...
LSTM多维预测数据|python
10-06
利用LSTM循环神经网络对多维数据进行预测，先对数据进行归一化处理，划分为训练集、测试集，用4个参数预测一个参数值
Python PX化工数据时间序列预测 LSTM GRU 预测5天后的数值 tensorflow KNN线性回归岭回归随机森林
01-27
Python PX化工数据时间序列预测 LSTM GRU 预测5天后的数值 各种排列组合的预测分别输出结果对比预测效果 2个自变量1个因变量 KNN 线性回归岭回归随机森林 early stoppping 输出csv numpy pandas sklearn tensorflow ...
VMD-LSTM时间序列预测（Python完整源码和数据）
05-06
在"VMD-LSTM.ipynb"文件中，包含了实现这个过程的完整Python源代码，包括数据读取、VMD实现、LSTM模型构建和训练、以及预测结果的可视化。通过阅读和理解这个代码，你可以深入学习如何将VMD与LSTM结合应用于实际问题...
爬虫学习——获得数据后的数据存储（csv & Excel 的写入与读取）
weixin_53919192的博客
04-21 5817
本文介绍了网页爬虫获取数据后数据存储两种方法，希望对你有帮助，加油，希望你我一同走进爬虫的世界~~
使用LSTM模型，以时间为顺序，对csv文件中的“收盘”这一列的数据进行预测，写出python代码...
weixin_35754676的博客
02-18 178
根据您提供的信息，可以使用以下python代码来对csv文件中“收盘”这一列的数据进行LSTM预测：# 导入相关库 import numpy as np import pandas as pd from keras.models import Sequential from keras.layers import LSTM from keras.layers import Dense# 导入数据 ...
Python 读取CSV文件，作散点图分析
congxinbaba的博客
07-10 2524
python 如何读取csv文件？如何作散点图？
python程序分析csv文件并绘制趋势图
gxs_hhxx的博客
02-18 2911
写了一个读取csv文件（第一列是时间，其余列是一些要分析记录变化趋势的数值），其中第一列的时间有进2k行而且不是标准的时间格式。 import matplotlib.pyplot as plt from numpy import genfromtxt import pandas as pd def draw(name,df,col): totalSeed = df.index.tolist() #...
爬取百度新闻写入CSV文件——Python爬虫
竹道生的博客
02-24 4861
在开始之前，我们先明确我们本次要爬什么内容： 接着，我们大概会用到以下几个知识点： re requests xpath csv 第一步，我们先将一些变量准备好 import requests import urllib.parse import os from lxml import etree import random import csv import re word="税" #转...
Python爬虫3.2 — csv用法教程
让编程改变世界
09-29 3660
Python爬虫3.2 — csv用法教程综述csv 介绍什么是csvcsv库的使用`reader()`读操作`DictReader()`读操作`writer()`写操作`DictWriter()`写操作总结其他博文链接 综述 本系列文档用于对Python爬虫技术的学习进行简单的教程讲解，巩固自己技术知识的同时，万一一不小心又正好对你有用那就更好了。 Python 版本是3.7.4 上一篇文章讲了...
使用python3爬取数据至csv
一直在路上的求知人。
09-20 4502
使用python3爬取数据至csv
爬虫之CSV文件存储
实践求真知
01-14 6569
一 点睛 CSV，全称为Comma-Separated Values，中文可以叫作逗号分隔值或字符分隔值，其文件以纯文本形式存储表格数据。该文件是一个字符序列，可以由任意数目的记录组成，记录间以某种换行符分隔。每条记录由字段组成，字段间的分隔符是其他字符或字符串，最常见的是逗号或制表符。不过所有记录都有完全相同的字段序列，相当于一个结构化表的纯文本形式。它比Excel文件更加简洁，XLS文本是电...
机器学习使用Python快速上手LSTM模型预测时间序列.pdf
最新发布
11-25
LSTM（长短期记忆网络）是一种递归神经网络，特别适用于处理时间序列数据，如预测连续变化的值，如商品销量或地下水位埋深。文章以一个实际案例为例，即通过分析煤矿监测井的地下水位数据，展示如何利用Python和LSTM...

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论 3
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

BRYTLEVSON

打赏的都是天使，创作的动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付：¥1
获取中
扫码支付

您的余额不足，请更换扫码支付或充值

打赏作者

实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值