Similar Libraries - Githubissues

simpeg / processing

software for time-series analysis, spectral analysis and transfer function computation from geophysical data

MIT License

6 stars 3 forks source link

Open lheagy opened 7 years ago

lheagy commented 7 years ago

Which libraries have similar functionality? What can we leverage? What can we look to for inspiration / ideas?

Obspy
- Does this already have time series analysis? What is the potential overlaps here?
Hdf5
- For storage
Mtpy (https://github.com/geophysics/mtpy)
Pandas
- E.g. single analysis, fft.
Astropy? (http://www.astropy.org/)
Scipy.signal
- Fft

kujaku11 commented 7 years ago

Scipy.signal also contains FIR and IIR filters that are useful for filtering time series as well as resampling, windowing, and FFT, probably one of the more useful packages.
Obspy has some filtering capabilities and spectral analysis, however they only calculate the power spectra which does not contain any phase information which is what we are most interested in.
Pandas is probably the most useful package that we can leverage from.
Statsmodel http://www.statsmodels.org/stable/index.html could also be a useful package for time series processing and statistical analysis.

thast commented 7 years ago

Just saw this package for signal processing, that will be part of SciPy 2017 tutorials: https://github.com/mwickert/scikit-dsp-comm

craigmillernz commented 7 years ago

GMT is being ported to python and has a bunch of very useful grid manipulation tools. http://gmt.soest.hawaii.edu/projects/gmt-python-api/wiki

This project has only just begun, but should be interesting as GMT is a solid set of tools.

lheagy commented 7 years ago

This is a seismic processing package out of UAlberta: https://github.com/SeismicJulia/Seismic.jl Some of the structure and organization may be applicable here too

thast commented 7 years ago

There were a tutorial on basic Times Series Processing with Pandas at SciPy2016: Youtube video of the tutorial: https://www.youtube.com/watch?v=JNfxr4BQrLk Github repository: https://github.com/AileenNielsen/TimeSeriesAnalysisWithPython

lheagy commented 7 years ago

xtensor might also be helpful (for code and or ideas): https://github.com/QuantStack/xtensor

ahartikainen commented 7 years ago

Hey, instead of pandas should we use xarray? It has the same indexing engine as the pandas.