discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: using gr-specest/gr-inspector


From: Aditya Arun Kumar
Subject: Re: using gr-specest/gr-inspector
Date: Fri, 14 May 2021 18:56:25 +0530

Thanks, Jeff and Daniel. I'll try to do them.



Sender notified by
Mailtrack 05/14/21, 06:56:02 PM

On Fri, May 14, 2021 at 6:41 PM Daniel Estévez <daniel@destevez.net> wrote:
Hi,

The documentation is here:

https://wiki.gnuradio.org/index.php/GNU_Radio_3.8_OOT_Module_Porting_Guide

In this case I think that only one block rather than a whole OOT needs
to be ported to GNU Radio 3.8, so basically the only changes that need
to be done is XML to YAML .grc files (gr_modtool helps with the
conversion) and python2 -> python3 in case it is a Python block.

In case of doubt or problems, it helps to see what people have done in
other OOTs. Most of them have maint-3.7 and maint-3.8 branches that you
can compare.

Best,

Dani.

El 14/5/21 a las 14:37, Jeff Long escribió:
> The general approach is to make a new OOT module for the version you
> want, using gr_modtool, then move code from the old module to the new
> module as needed.
>
> On Fri, May 14, 2021 at 7:25 AM Aditya Arun Kumar
> <adityaarunkumarphi@gmail.com <mailto:adityaarunkumarphi@gmail.com>> wrote:
>
>     Thanks Marcus. I was also thinking the same thing. Can you please
>     link me to some porting guides, I am new to this and the last time I
>     tried I failed miserably.
>
>
>
>     Mailtrack
>     <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>
>       Sender notified by
>     Mailtrack
>     <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>
>     05/14/21, 04:51:09 PM     
>
>
>     On Fri, May 14, 2021 at 3:21 PM Marcus Müller <mueller@kit.edu
>     <mailto:mueller@kit.edu>> wrote:
>
>         Hi Aditya!
>
>         I think the most straightforward way would be to port the AMC
>         block to 3.8. I wasn't aware
>         of it existing only in maint-3.7!
>
>         Best regards,
>         Marcus
>
>         On 14.05.21 00:50, Aditya Arun Kumar wrote:
>          > Hi Marcus,
>          > Thanks for the reply. The reason I am sticking with GRC 3.7
>         branch is that the AMC block
>          > is present only in the gr-inspector 3.7 branch, which
>         requires a CNN graph for input, I
>          > have no issues in running the 3.8 branches of the OOT. So, if
>         this is the case is there a
>          > way I can import a TensorFlow graph model into GRC3.8
>         because the inspector branch really
>          > says it is not present.
>          >
>          >
>          >
>          > Mailtrack
>          >
>         <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&
>         <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>>
>          >       Sender notified by
>          > Mailtrack
>          >
>         <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&
>         <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>>
>          > 05/14/21, 04:14:53 AM
>          >
>          >
>          > On Fri, May 14, 2021 at 4:08 AM Marcus Müller
>         <mmueller@gnuradio.org <mailto:mmueller@gnuradio.org>
>          > <mailto:mmueller@gnuradio.org
>         <mailto:mmueller@gnuradio.org>>> wrote:
>          >
>          >     Hi Aditya,
>          >
>          >     you shouldn't use GNU Radio 3.7 for new things in 2021,
>         it's a dead end; we don't develop
>          >     this any further, and it will not work on newer machines;
>         there's nothing we can do about
>          >     that.
>          >
>          >     Both gr-inspector as well as gr-specest work with GNU
>         Radio 3.8, if you check out their
>          >     maint-3.8 branch.
>          >
>          >     GNU Radio 3.8 is easily installable on any reasonably
>         modern operating system, so that's
>          >     what you should do.
>          >
>          >     Neither of this has anything to do with Neural Networks,
>         so you're on your own there.
>          >
>          >     Best regards,
>          >     Marcus
>          >
>          >     On 13.05.21 23:38, Aditya Arun Kumar wrote:
>          >     > Hello,
>          >     > I am working on AMC(Automatic Modulation
>         Classification) using GRC. After some digging
>          >     > around I found that gr-inspector along with gr-specest
>         will work for this. I also found
>          >     > that while using gr-inspector I need to check out the
>         maint-3.7 branch (gr-inspector for
>          >     > maint3.7 says that a minimum of GRC 3.7.9.3 is
>         required) for me to load the tf
>          >     models. Now
>          >     > I face problems in 2 places,
>          >     >
>          >     >
>          >     > 1. gr-specest is not really available for 3.7 at
>         least in the kit-cell branch, but
>          >     > available here
>         <https://github.com/robertwatsonbath/gr-specest-3.7
>         <https://github.com/robertwatsonbath/gr-specest-3.7>
>          >     <https://github.com/robertwatsonbath/gr-specest-3.7
>         <https://github.com/robertwatsonbath/gr-specest-3.7>>>. So I
>         install that and
>          >     > try to run the example flowgraph I get "Module not
>         found: _swig_specest", the same
>          >     > happens when I try to do it from the terminal as well.
>         After this, I found this
>          >     > <https://signalsintelligence.wordpress.com/
>         <https://signalsintelligence.wordpress.com/>
>          >     <https://signalsintelligence.wordpress.com/
>         <https://signalsintelligence.wordpress.com/>>> blog that talks
>         about the problems I have,
>          >     > but the problem is I cannot find any of the links in
>         git-repo as most of them are dead.
>          >     > Now my question in this place is there a specific
>         gr-specest branch for GR3.7.
>          >     > 2. The second part of the problem is where can I find a
>         pre-trained CNN model that I can
>          >     > import and use (I am still learning the NN parts). If I
>         cannot find is there a
>          >     > well-documented t project that I can use to generate my
>         CNN graph that I can import
>          >     as AMC
>          >     > into gr-inspector tensor-flow block.
>          >     >
>          >     > But whatever it looks like I am going to need a CNN for
>         mod rec at the moment, but what
>          >     > puzzles me is I cannot find gr-specest anywhere just to
>         understand some 3D visualization
>          >     > as well for 3.7. Any help in this aspect will be really
>         helpful.
>          >     >
>          >     > --
>          >     > S. Aditya Arun Kumar
>          >     >
>          >     >
>          >     > Mailtrack
>          >     >
>          >   
>           <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5& <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>
>          >   
>           <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5& <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>>>
>          >     >       Sender notified by
>          >     > Mailtrack
>          >     >
>          >   
>           <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5& <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>
>          >   
>           <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5& <https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality5&>>>
>          >     > 05/14/21, 02:54:51 AM
>          >     >
>          >
>          >
>          >
>          > --
>          > S. Aditya Arun Kumar
>          > +919123517465
>
>
>
>     --
>     S. Aditya Arun Kumar
>     +919123517465
>




--
S. Aditya Arun Kumar
+919123517465

reply via email to

[Prev in Thread] Current Thread [Next in Thread]