search

laszukdawid / PyEMD

Python implementation of Empirical Mode Decompoisition (EMD) method
https://pyemd.readthedocs.io/
Apache License 2.0
867 stars 224 forks source link

Add multidimens support #4

Closed laszukdawid closed 2 years ago

laszukdawid commented 7 years ago

It seems that there is a need for multidimensional support. This should be included soon.

alexanderhudym commented 6 years ago

I think this would be fine if i could do the decomposition for pandas date frames at once

laszukdawid commented 6 years ago

Can you specify what exactly you mean? Is a single column a (time) series?

On Sun, May 6, 2018, 06:33 Alexander notifications@github.com wrote:

I think this would be fine if i could do the decomposition for pandas date frames at once

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/laszukdawid/PyEMD/issues/4#issuecomment-386879755, or mute the thread https://github.com/notifications/unsubscribe-auth/AK7Vqx-36jcVl_o2_wlauU4mwWNsu3T3ks5tvvuMgaJpZM4Nd-LD .

alexanderhudym commented 6 years ago

@laszukdawid something like this df = pd.read_csv('btc_usd.csv') //columns: open, close, high, volume, etc memd = MEMD() //multidimentional imfs = memd(df) or imfs = memd(df.values)

laszukdawid commented 6 years ago

@alexanderhudym df.values returns np.ndarray so that should be out of box for the future implementation. Although, there's less and less time, and more and more other projects so the priority of MEMD is going down. Still haven't seen any good article that explains how MEMD could work.

laszukdawid commented 2 years ago

No way this is going to happen. Open for over 4 years? Past me didn't want; current me doesn't want... let me save future me some thinking.

Closing.