wavelet packet decomposition
SantiagoLunaRomero opened this issue · comments
if you compare your code with the funtion "wpdec()" in matlab.
What is the diference?
Hi, I'm using this library to implement a project that once was implemented in Matlab. I'd validated it uses but it has a few differences:
Matlab "appcoef" returns the recomposition of "wavedec" decomposition, using the Wavelet parameters you choosed. When you return your "appcoef" results, it will show (N + 5) samples, 'N' as 1/(2^s) of total samples you enter in "wavedec" function and 's' the scale order you choosed to recompose.
Using this library, when you recompose your signal, it will return the same number of samples, but in fact, only the N/(2^s) is the recomposition data.
I don't know if this library has an option to fix it automatically, since I just began to use Wavelet and this library.
Hope it helps.
Regards.
Hi,
Can you send me the code so I can replicate the issue ? This could be
because of the type of extension choosen as symmetric and preiodic
extensions will return different size.
Rafat
…On 2/23/17, fredpreve ***@***.***> wrote:
Hi, I'm using this library to implement a project that once was implemented
in Matlab. I'd validated it uses but it has a few differences:
Matlab "appcoef" returns the recomposition of "wavedec" decomposition,
using the Wavelet parameters you choosed. When you return your "appcoef"
results, it will show (N + 5) samples, 'N' as 1/(2^s) of total samples you
enter in "wavedec" function and 's' the scale order you choosed to
recompose.
Using this library, when you recompose your signal, it will return the same
number of samples, but in fact, only the N/(2^s) is the recomposition data.
I don't know if this library has an option to fix it automatically, since I
just began to use Wavelet and this library.
Hope it helps.
Regards.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
#5 (comment)
Hello. I appreciate that you answer me! By the way, thank you very much for
the library, It helped me a lot.
Yes, I used symmetric config. I took directly from the dwt example. I knew
I was doing something wrong.
Again, thank you very much!
Regards!
Em 24 de fev de 2017 6:49 AM, "Rafat Hussain" <notifications@github.com>
escreveu:
… Hi,
Can you send me the code so I can replicate the issue ? This could be
because of the type of extension choosen as symmetric and preiodic
extensions will return different size.
Rafat
On 2/23/17, fredpreve ***@***.***> wrote:
> Hi, I'm using this library to implement a project that once was
implemented
> in Matlab. I'd validated it uses but it has a few differences:
>
> Matlab "appcoef" returns the recomposition of "wavedec" decomposition,
> using the Wavelet parameters you choosed. When you return your "appcoef"
> results, it will show (N + 5) samples, 'N' as 1/(2^s) of total samples
you
> enter in "wavedec" function and 's' the scale order you choosed to
> recompose.
>
> Using this library, when you recompose your signal, it will return the
same
> number of samples, but in fact, only the N/(2^s) is the recomposition
data.
>
> I don't know if this library has an option to fix it automatically,
since I
> just began to use Wavelet and this library.
>
> Hope it helps.
> Regards.
>
> --
> You are receiving this because you are subscribed to this thread.
> Reply to this email directly or view it on GitHub:
> #5 (comment)
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#5 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/ARGVxm3-PH2I5uE3WjvLF0_b-y9HRpgyks5rfqergaJpZM4LaSPn>
.
The dwttest result is different from pywt result
You should get the same results if you are using the same wavelet, levels
of decomposition and signal extension.
Figure 2) is the actual DWT detail coefficients.
Figure 3) is the actual DWT approximation coefficients.
The first figure is the reconstructed approximation and detail coefficients
over a single branch so that they each have the same length and can be used
for signal analysis. You should be able to view the actual coefficients
using the dropdown menu of the graph. For example, "Details at level 1" etc.
Rafat
<https://www.avast.com/en-in/recommend?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=default3&tag=8aa13abc-7b06-4499-af86-2eee179ffd91>
I’m
protected online with Avast Free Antivirus. Get it here — it’s free forever.
<https://www.avast.com/en-in/recommend?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=default3&tag=8aa13abc-7b06-4499-af86-2eee179ffd91>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
…On Mon, Jul 1, 2019, 13:48 zhanghonglang ***@***.***> wrote:
The dwttest result is different from pywt result
[image: image]
<https://user-images.githubusercontent.com/13651762/60421149-d13d0d00-9c1b-11e9-8eb8-ce22708e7346.png>
[image: image]
<https://user-images.githubusercontent.com/13651762/60421010-8622fa00-9c1b-11e9-8747-3edd9e25974b.png>
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#5?email_source=notifications&email_token=AAI3RO27MLIX6KNB4LCT7NLP5G4WJA5CNFSM4C3JEPT2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODY5LUYQ#issuecomment-507165282>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAI3RO6DJGXIJNMUZJMRQ23P5G4WJANCNFSM4C3JEPTQ>
.