Ok. I've found a way (to copy an fft object) :
lfft = len(fftobj)
fftobj_copy = FFT(lfft)
for i in range(lfft):
fftobj_copy[i]=fftobj[i]
If you see things which I could / should do another way in my code,
feel free to post your suggestions!
On Dec 18, 11:45 am, Maxim <maxim.courno...@gmail.com> wrote:
> Hello!
>
> I am working with the builtin FFT Fast Fourier Transform of Sagemath,
> and have coded (based on the work of P. Lutus here
> :http://vps.arachnoid.com/sage/fourier.html) a function that takes an
> fftobject to plot the spectrum of a function.
>
> My problem is that when I pass the fftobject, I don't know how to make
> a local copy of this object inside the scope of my function. The
> result if that after my first call of this function, the fftobject
> gets modified (globally) when I run fftobject.fordward_transform() on
> it. This is not what I want, because I want to reuse this same object
> unchanged further in my code.
>
> I'm new at Python, so I do not know a lot about this. I have first
> tried to to something as :
> fftobject_copy = fftobject
> But it would just create a pointer to fftobject, and still allow the
> function to modify its content. I tried to use the list copy :
> fft_object_copy = fftobject[:]
> But it would loose it's FFT specific functions, such as in
> "fftobject.forward_transform()".
>
> So, how can I make a local copy of an FFT object? An example of my
> code is following.
> Many thanks!
>
> ## My functions declaration. Notice the fft_unilateral(fftobj)
> function, where I'm trying to make a local copy of
> ## fftobj.
> # magnitude of 2-component cartesian vector
> def mag(x):
> return sqrt(x[0]^2+x[1]^2)
>
> def nextpow2(i):
> n = 2
> while n < i: n = n * 2
> return n
>
> def fft_unilateral(fftobj):
> # fftobj_copy = fftobj[:] <== Trying to create a local copy of
> fftobj here!
> fftobj_copy = fftobj # <== fftobj_copy is not a copy but a
> pointer to fftobj!
> fftobj_copy.forward_transform()
> lfft = len(fftobj_copy)
> dt = 1.0/lfft
> list = map(lambda x:(2*mag(x))*dt,fftobj_copy[:lfft/2])
> list[0]=list[0]/2
> return list
>
> # frequency unilateral domain plot
> def fft_plot(fftobj,line_color='blue',labels=
> ('Frequence','Amplitude')):
> list = fft_unilateral(fftobj)
> return list_plot
> (list,rgbcolor=line_color,plotjoined=True,axes_labels=labels)
>
> # Power Spectrum Density plot
> def psd_plot(fftobj,line_color='blue',labels=
> ('Frequence','Amplitude')):
> list = fft_unilateral(fftobj)
> for i in range(len(list)):
> list[i]=list[i]^2
> return list_plot
> (list,rgbcolor=line_color,plotjoined=True,axes_labels=labels)
>
> ## Begin main program
> # Define time domain function (AM without overmodulation here)
> m(t)= cos(2*pi*100*t)
> p(t)= cos(2*pi*1000*t)
> AM1(t)=(1 + 0.5*m(t))*p(t)
>
> # Create
> samples=4000
> lfft=nextpow2(samples)
> # samples = sampling frequency
> # Create fft object
> fft_a=FFT(lfft)
> # Fill fftobject with time domain data
> for t in range(lfft):
> fft_a[t] = AM1(t/lfft)
>
> ## Finally, ready to fire it up.
> # If you have succeded to create a local copy in the fft_unilateral
> function, then the two following PSDs are
> # identical. If not, then they differ.
> fft_plot(fft_a,line_color='blue',labels=('$f$ (Hz)','$|\Phi(f)|$'))
>
> # Copy to new cell
> fft_plot(fft_a,line_color='blue',labels=('$f$ (Hz)','$|\Phi(f)|$'))
--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
