Dear SymPy Community,
I just wanted to share with you, that the TrendPy Project about Time Series Regressions with Python
is now uploaded to pip (as trendpy2, i.e. pip install trendpy2).
Reminder:
The trendpy2 package makes it easy to approximate time series regressions in a determinstic way. The following trends are supported:
linear
polynomial
exponential
trigonometric
“free”, for max. three parameters, e.g. (the intial guess for a, b, c is 1.)
A standalone feature of the trendpy2 package is, that it combines least-squares approaches, Fourier analysis approaches, numerical Python packages as Numpy and Scipy and the symbolic Python package Sympy for time series regressions.
SymPy is makes it possible to easlily implement a "free" regression approach.
A streamlit web app is now released, additionally to the voila web app. It can be tried out using this link (also linked in the github page)
https://zolabar-trendpy-trendpy2-app-kfqshb.streamlit.app/Testdata in
Enjoy! Feedback is welcome ;)
Regards,
Zoufiné