我正在尝试创建一个ARIMA模型进行预测但是我收到了一个错误
“输入类型不支持ufunc'isfinite',根据强制规则''safe'',输入无法安全地强制转换为任何支持的类型
我正在尝试运行代码:
Data= pd.read_csv(xcjs_2018.csv', low_memory=False)
Data['day'] = pd.to_datetime(Data['day'])
from plotly.plotly import plot_mpl from statsmodels.tsa.seasonal
import seasonal_decompose
result = seasonal_decompose(Data, model= 'multiplicative')
fig = result.plot()
plot_mpl(fig)
我的Dataframe看起来像这样:
Date Name COUNT
2018-08-03 XCJS_22 199
2018-08-04 XCJS_22 200
2018-08-06 XCJS_22 151
2018-08-07 XCJS_22 159
2018-08-08 XCJS_22 451
2018-08-15 XCJS_22 217
2018-08-20 XCJS_22 389
Traceback error message有谁知道如何帮助?
尝试将日期列创建为索引,并使用freq = 1绘制Data ['Count']列。
#reads csv file
Data= pd.read_csv(xcjs_2018.csv', low_memory=False)
Data['Date'] = pd.to_datetime(Data['Date'])
Data = Data.set_index('Date')
#Plot it
from plotly.plotly import plot_mpl
from statsmodels.tsa.seasonal import seasonal_decompose
result = seasonal_decompose(Data.COUNT, model= 'multiplicative', freq=1)
fig = result.plot()
plot_mpl(fig)