# 使用Prophet预测manning未来365天的页面流量

# 从2007年12月10日开始
import pandas as pd
from fbprophet import Prophet
import matplotlib.pyplot as plt
%matplotlib inline

# 读入数据集
df = pd.read_csv('./manning.csv')
#print(df.head())
#print(df.tail())
# 拟合模型
model = Prophet()
model.fit(df)

# 构建待预测日期数据框，periods = 365 代表除历史数据的日期外再往后推 365 天
future = model.make_future_dataframe(periods=365)
#print(future.tail())

# 预测数据集
forecast = model.predict(future)
#print(forecast.columns)
print(forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail())

# 展示预测结果
model.plot(forecast)
plt.show()

#print(forecast)
forecast.tail()

# 预测的成分分析绘图，展示预测中的趋势、周效应和年度效应
model.plot_components(forecast)
print(forecast.columns)

# 饱和增长
df['cap'] = 8.5
m = Prophet(growth='logistic')
m.fit(df)

# 预测未来 3 年的数据
future = m.make_future_dataframe(periods=1826)
# 将未来的承载能力设定得和历史数据一样
future['cap'] = 8.5
fcst = m.predict(future)
fig = m.plot(fcst)
print(future)

# 预测饱和减少
df['y'] = 10 - df['y']
df['cap'] = 6
# 设置下限
df['floor'] = 1.5
future['cap'] = 6
future['floor'] = 1.5
m = Prophet(growth='logistic')
m.fit(df)
fcst = m.predict(future)
fig = m.plot(fcst)

from fbprophet.plot import add_changepoints_to_plot
fig = m.plot(forecast)
# plt.gca()获得当前的Axes对象ax
a = add_changepoints_to_plot(fig.gca(), m, forecast)

# 指定突变点的位置
m = Prophet(changepoints=['2014-01-01'])
m.fit(df)
future = m.make_future_dataframe(periods=365)
forecast = m.predict(future)
m.plot(forecast);

# 对节假日建模
playoffs = pd.DataFrame({
  'holiday': 'playoff',
  'ds': pd.to_datetime(['2008-01-13', '2009-01-03', '2010-01-16',
                        '2010-01-24', '2010-02-07', '2011-01-08',
                        '2013-01-12', '2014-01-12', '2014-01-19',
                        '2014-02-02', '2015-01-11', '2016-01-17',
                        '2016-01-24', '2016-02-07']),
  'lower_window': 0,
  'upper_window': 1,
})
superbowls = pd.DataFrame({
  'holiday': 'superbowl',
  'ds': pd.to_datetime(['2010-02-07', '2014-02-02', '2016-02-07']),
  'lower_window': 0,
  'upper_window': 1,
})
holidays = pd.concat((playoffs, superbowls))
m = Prophet(holidays=holidays)
m.fit(df)
future = m.make_future_dataframe(periods=365)
forecast = m.predict(future)
#print(forecast)
# 通过 forecast 数据框，展示节假日效应
forecast[(forecast['playoff'] + forecast['superbowl']).abs() > 0][['ds', 'playoff', 'superbowl']][-10:]
# 可以使用 plot_forecast_component（从fbprophet.plot导入）来画出独立的节假日的成分

from fbprophet.plot import plot_forecast_component
m.plot_forecast_component(forecast, 'superbowl')

# 预测的成分分析绘图，展示预测中的趋势、周效应和年度效应
model.plot_components(forecast)
print(forecast.columns)