Let's say I have two structured arrays with dtypes as per below
getdat.dtype
dtype([('Tstamp', '|O4'), ('Vf', 'f4'), ('Vq', 'f4'), ('Sf', 'f4'),
('Sq', 'f4'), ('Bt', 'f4'), ('SPL', 'f4')])
out.dtype
dtype([('Viscosity_cSt', 'f4'), ('Density_kgm3', 'f4'), ('GVF', 'f4')])
Then merging this per
Hi group,
is there a faster way to calculate the
matrix exponential for different t's
than this:
def sol_matexp(A, tlist, y0):
w, v = np.linalg.eig(A)
out = np.zeros((tlist.size, y0.size))
for i, t in enumerate(tlist):
sol_t =
On Mon, Jan 28, 2013 at 5:31 PM, Till Stensitzki mail.t...@gmx.de wrote:
Hi group,
is there a faster way to calculate the
matrix exponential for different t's
than this:
def sol_matexp(A, tlist, y0):
w, v = np.linalg.eig(A)
out = np.zeros((tlist.size, y0.size))
for i, t in
I did not try it, but I assume that you can build a stack of diagonal matrices
as a MxNxN array and use tensordot with the matrix v (and it's inverse). The
trivial way to accelerate the loop is to calculate in inverse of v before the
loop.
Nadav
Hi,
Le 28/01/2013 17:31, Till Stensitzki a écrit :
This is the calculates exp(-Kt).dot(y0) for a list a ts.
If your time vector ts is *regularly* discretized with a timestep h, you
could try an iterative computation
I would (roughly) write this as :
Ah = np.expm(A*h) # or use the
Thanks for hints so far, i am especially searching
for a way to get rid of the t loop. Making a NxMxM
Matrix is quite memory inefficient in my case (N M).
On way would be just use cython, but i think this problem
common enough to have a solution into scipy.
(Solution of a simple compartment
Hi,
Le 28/01/2013 18:14, Till Stensitzki a écrit :
On way would be just use cython, but i think this problem
common enough to have a solution into scipy.
(Solution of a simple compartment model.)
I see the solution you propose as a specialized ODE solver for linear
systems.
Then, what about
Hello,
I have embedded python/numpy scripts in an application that runs in
parallel.
But the python code is always invoked on the master node.
So it could be assumed that at some point, there could be multiple
instances of script being invoked and run.
I have commented out import numpy part to
Please ignore the previous message. I have done some testing and found it
to be running on a client node instead of the master node.
The problem might be because node2, does not have numpy installed. Thanks.
___
NumPy-Discussion mailing list
Hi,
I have a somewhat convoluted N-dimensional array that contains information of a
set of experiments.
The last dimension has as many entries as iterations in the experiment (an
iterative application), and the penultimate dimension has as many entries as
times I have run that experiment; the
I discovered this from C via the PyArray_FromAny function, but here it
is in Python:
asarray(None,dtype=float)
array(nan)
Is this expected or documented behavior? It seems quite unintuitive
and surprising that this wouldn't throw an exception.
Is there a way to disable this behavior
import numpy as np
np.double(None)
nan
On Mon, Jan 28, 2013 at 3:48 PM, Geoffrey Irving irv...@naml.us wrote:
I discovered this from C via the PyArray_FromAny function, but here it
is in Python:
asarray(None,dtype=float)
array(nan)
Is this expected or documented behavior?
For comparison:
float32(None)
nan
float(None)
Traceback (most recent call last):
File stdin, line 1, in module
TypeError: float() argument must be a string or a number
On Mon, Jan 28, 2013 at 5:09 PM, Bradley M. Froehle
brad.froe...@gmail.com wrote:
import numpy as np
np.double(None)
nan
13 matches
Mail list logo