foundational timeseries model Regressor

Switched to a regression function based on the MomentFM framework. Reconstruction/Regression can be applied on raw, irrgeularly sampled data, containing arbitrary NaN values.

data = pd.read_csv('/resources/data/hydro_data.csv', parse_dates=['Timestamp'], index_col=0, nrows=10000)
qc = saqc.SaQC(data)
qc = qc.fitFMmoment(['sac254_raw'], target=['sac254_ttm'])
qc.plot(['sac254_raw','sac254_ttm'], mode='oneplot')

MomentFM requirements have to be installed as extras (saqc[FM])

added category: algorithms label

requested review from @schaefed

assigned to @luenensc

added 1 commit

• b9f8851f - moved ttm import

Compare with previous version

added 1 commit

• 5f3f74e0 - moved tsfm imports again

Compare with previous version

changed the description

added 2 commits

• 59fbc2ea - added embedded regression
• 5221e824 - black/isort

Compare with previous version

changed the description

added 2 commits

• 8b99583f - bfx
• 8dd995d3 - ..

Compare with previous version