网站建设 福田,顶易云外贸软件,我的百度账号,福田网站建设运营费用时间序列预测#xff08;一#xff09;基于Prophet的销售额预测 小O#xff1a;小H#xff0c;有没有什么方法能快速的预测下未来的销售额啊 小H#xff1a;Facebook曾经开源了一款时间序列预测算法fbprophet#xff0c;简单又快速#xff5e; 传统的时间序列算法很多一基于Prophet的销售额预测 小O小H有没有什么方法能快速的预测下未来的销售额啊 小HFacebook曾经开源了一款时间序列预测算法fbprophet简单又快速 传统的时间序列算法很多例如AR、MA、ARIMA等对于非专业人员来说显得很难上手。而Prophet相对来说就友好多了而且预测效果又很不错所以用它来预测时间序列数据再适合不过了。本文主要参考基于facebook的时间序列预测框架prophet的实战应用。 Prophet的安装需要先安装pystan conda install pystan # 终端上安装需要执行procced选择ypip install fbprophet数据探索
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import seaborn as sns
from fbprophet import Prophet
from sklearn.metrics import mean_squared_error
from math import sqrt
import datetime
from xgboost import XGBRegressor
from sklearn.metrics import explained_variance_score, mean_absolute_error, \
mean_squared_error, r2_score # 批量导入指标算法from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import LSTM
from tensorflow.keras.layers import Dense, Dropout
from sklearn.preprocessing import MinMaxScaler
from keras.wrappers.scikit_learn import KerasRegressor
from sklearn.model_selection import GridSearchCV以下数据如果有需要的同学可关注公众号HsuHeinrich回复【数据挖掘-时间序列01】自动获取 # 读取数据
raw_data pd.read_csv(train.csv)
raw_data.head()# 转化为日期
raw_data[datetime] raw_data[datetime].apply(pd.to_datetime)# 查看历史销售趋势
plt.figure(figsize (15,8))
sns.lineplot(x datetime, y count, data raw_data, err_styleNone)
plt.show()特征工程
# 构造prophet需要的ds/y数据
df_model raw_data[[datetime, count]].rename(columns {datetime: ds,count: y})模型拟合
# 模型拟合
model_fb Prophet(interval_width 0.95).fit(df_model)
# 构造预测日期
future_dates model_fb.make_future_dataframe(periods 100, freqH)
# 预测结果
forecast model_fb.predict(future_dates)INFO:fbprophet:Disabling yearly seasonality. Run prophet with yearly_seasonalityTrue to override this.# 预测最后几周的日期
forecast[[ds, yhat, yhat_lower, yhat_upper]].tail()结果展示
# 观察预测效果
model_fb.plot(forecast);# 观察趋势因素
model_fb.plot_components(forecast);# 模型评估 MSE
metric_df forecast.set_index(ds)[[yhat]].join(df_model.set_index(ds).y).reset_index()
metric_df.dropna(inplaceTrue)
error mean_squared_error(metric_df.y, metric_df.yhat)
print(The MSE is {}. format(error))The MSE is 12492.842870220222添加假期因素
# 定义假期因素
def is_school_holiday_season(ds): date pd.to_datetime(ds)starts datetime.date(date.year, 7, 1)ends datetime.date(date.year, 9, 9)return starts date.to_pydatetime().date() endsdf_model[school_holiday_season] df_model[ds].apply(is_school_holiday_season)
df_model[not_school_holiday_season] ~df_model[ds].apply(is_school_holiday_season)
model_fb Prophet(interval_width0.95)# 添加假期因素
model_fb.add_seasonality(nameschool_holiday_season, period365, fourier_order3, condition_nameschool_holiday_season)
model_fb.add_seasonality(namenot_school_holiday_season, period365, fourier_order3, condition_namenot_school_holiday_season)
model_fb.fit(df_model)INFO:fbprophet:Disabling yearly seasonality. Run prophet with yearly_seasonalityTrue to override this.fbprophet.forecaster.Prophet at 0x7ff4e48833d0# 构造日期
future_dates model_fb.make_future_dataframe(periods100, freqH)
future_dates[school_holiday_season] future_dates[ds].apply(is_school_holiday_season)
future_dates[not_school_holiday_season] ~future_dates[ds].apply(is_school_holiday_season)# 预测
forecast model_fb.predict(future_dates)plt.figure(figsize(10, 5))
model_fb.plot(forecast);Figure size 720x360 with 0 Axes# 观察趋势因素
model_fb.plot_components(forecast); # 模型评估 MSE
metric_df forecast.set_index(ds)[[yhat]].join(df_model.set_index(ds).y).reset_index()
metric_df.dropna(inplaceTrue)
error mean_squared_error(metric_df.y, metric_df.yhat)
print(The MSE is {}. format(error))The MSE is 12431.431390456968添加假期因素后预测上没有提升。这里只是介绍如何增加自定义趋势因素而已所以没有提升在预期之内 总结
当你只需要预测数据时只需简单的两列dsy即可整个预测过程简单易上手
共勉
