Eheran1
commented
2 years ago Also, this Line here is wrong:
Fs = [np.array([[1., dt], [0, 1]] for dt in dts]
Not only is there a missing ")", but also it throws errors:
VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.
Fs = [np.array([[1., dt], [0., 1.]], dtype=object) for dt in dts] Input In [2] in <module>
runfile('C:/Users/me/Desktop/Code/KF-3.12-tempdaten.py', wdir='C:/Users/me/Desktop/Code')
File C:\ProgramData\Anaconda3\lib\site-packages\debugpy\_vendored\pydevd\_pydev_bundle\pydev_umd.py:167 in runfile
execfile(filename, namespace)
File C:\ProgramData\Anaconda3\lib\site-packages\debugpy\_vendored\pydevd\_pydev_imps\_pydev_execfile.py:25 in execfile
exec(compile(contents + "\n", file, 'exec'), glob, loc)
File C:/Users/me/Desktop/Code/KF-3.12-tempdaten.py:154 in <module>
(mu, cov, _, _) = f.batch_filter(zs, Fs=Fs)
File C:\ProgramData\Anaconda3\lib\site-packages\filterpy\kalman\kalman_filter.py:911 in batch_filter
self.update(z, R=R, H=H)
File C:\ProgramData\Anaconda3\lib\site-packages\filterpy\kalman\kalman_filter.py:545 in update
self.SI = self.inv(self.S)
File <__array_function__ internals>:5 in inv
File C:\ProgramData\Anaconda3\lib\site-packages\numpy\linalg\linalg.py:545 in inv
ainv = _umath_linalg.inv(a, signature=signature, extobj=extobj)
UFuncTypeError: Cannot cast ufunc 'inv' input from dtype('O') to dtype('float64') with casting rule 'same_kind'This also happens with:
Fs = [np.array([[1., dt], [0., 1.]], dtype=object) for dt in dts]
What should Fs look like to get it working? I have a np.array of my dts. Ah never mind, like this:
Fs = []
for t in range(0, len(mytime)):
Fs.append(np.array([[1., dts[t], [0., 1.]]))
Ms, _, _, _ = f.rts_smoother(mu, cov, Fs=Fs)
In kalman_filter.py there is a code-block in the docstring reading:
However, rts_smoother returns 4 lists, not 3.
Also, get rid of those silly parens on the tuples!