ValueError: Time series must have a monotonic time index. about adtk

[tailaiw]

For your input s, what does s.index.is_monotonic give you?

[bigbprp]

I am getting false for s.index.is_monotonic.. not sure.. what am I missing... for details review below screen capture...

[tailaiw]

That means the Series index is not monotonic. You need to make sure your Series index is DatetimeIndex and it is monotonic. adtk.data.validate_series may be helpful to you.

[bigbprp]

Thank you for your inputs.. my date column has duplicate as it's without timestamp... I mean without hours or minutes or seconds...

As per business requirements that is a valid scenario... also anomaly detection is on monthly transaction basis so I am confused how to proceed... Any help is highly appreciated

[bigbprp]

@tailaiw, thank you very much for guidance.. .I just created a new Data-frame with group-by on dates & started analyzing... appreciate your timely response... so taking a small break helped..