FateMurphy / CEEMDAN_LSTM

CEEMDAN_LSTM is a Python project for decomposition-integration forecasting models based on EMD methods and LSTM.
MIT License
205 stars 42 forks source link

请求大佬帮助!!! #25

Open ScaleZhang opened 4 months ago

ScaleZhang commented 4 months ago

按照第一种方法pip ceemdan-lstm之后,直接调用了import CEEMDAN_LSTM as cl cl.quick_keras_predict(data=None) 但出现了这样的问题 求解答!!! ppp

FateMurphy commented 4 months ago

感谢来信。图中显示的是图形执行错误。你可以先试试把plot和save的路径更改。

Regards, Feite ZHOU, @.***

------------------ 原始邮件 ------------------ 发件人: ScaleZhang @.> 发送时间: 2024年2月23日 00:33 收件人: FateMurphy/CEEMDAN_LSTM @.> 抄送: Subscribed @.***> 主题: Re: [FateMurphy/CEEMDAN_LSTM] 请求大佬帮助!!! (Issue #25)

按照第一种方法pip ceemdan-lstm之后,直接调用了import CEEMDAN_LSTM as cl cl.quick_keras_predict(data=None) 但出现了这样的问题 求解答!!! ppp.jpg (view on web)

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>

gaochenxiai commented 4 months ago

屏幕截图 2024-03-01 194558 大佬请教一下,我把LSTM换成了bi -LSTM,效果变差了,会是哪里的原因呢

FateMurphy commented 4 months ago

我猜测应该是数据的问题,BiLSTM对于较短的序列和简单的模式,可能会出现过拟合的情况。

ScaleZhang commented 3 months ago

谢谢大佬回复!我在colab上跑就好了,但是我又有了新的问题,1.您的程序我是在colab上运行的,首先用single lstm跑了一次,结果26s就完成了,和您论文所描述的400多s不符合,我就在想是不是因为epoch设置的原因,我首先我看您的epoch设置的是100,我直接在改为了1000,然后重新跑了一下,发现还是很快就跑完了,不知道是怎么回事 2.对于Ensemble forecasting framework和Respective forecast预测都不需要二次分解,但是在调用前的kr里设置的参数有重分解等,我该怎么设置才能实现基本的ensemble forecast呢? 还望大佬解答,多谢!!!

------------------ 原始邮件 ------------------ 发件人: "FateMurphy/CEEMDAN_LSTM" @.>; 发送时间: 2024年2月29日(星期四) 中午12:07 @.>; @.**@.>; 主题: Re: [FateMurphy/CEEMDAN_LSTM] 请求大佬帮助!!! (Issue #25)

感谢来信。图中显示的是图形执行错误。你可以先试试把plot和save的路径更改。

Regards, Feite ZHOU, @.***

------------------ 原始邮件 ------------------ 发件人: ScaleZhang @.> 发送时间: 2024年2月23日 00:33 收件人: FateMurphy/CEEMDAN_LSTM @.> 抄送: Subscribed @.***> 主题: Re: [FateMurphy/CEEMDAN_LSTM] 请求大佬帮助!!! (Issue #25)

按照第一种方法pip ceemdan-lstm之后,直接调用了import CEEMDAN_LSTM as cl cl.quick_keras_predict(data=None) 但出现了这样的问题 求解答!!! ppp.jpg (view on web)

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.> — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.>

FateMurphy commented 3 months ago

1、论文用的是1060单显卡和旧版本的CUDA环境,和colab时间差距大很正常。 2、输入REDECOM_LIST=None,则不再执行二次分解。输入INTE_LIST=None,则不再执行子序列整合。