gnuastro-devel
[Top][All Lists]
Advanced

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

[task #16362] Library: The pooling program for the Gnuastro


From: Faezeh Bidjarchian
Subject: [task #16362] Library: The pooling program for the Gnuastro
Date: Thu, 4 May 2023 16:43:59 -0400 (EDT)

URL:
  <https://savannah.gnu.org/task/?16362>

                 Summary: Library: The pooling program for the Gnuastro 
                   Group: GNU Astronomy Utilities
               Submitter: fbidjarchian
               Submitted: Thu 04 May 2023 08:43:58 PM UTC
         Should Start On: Thu 04 May 2023 12:00:00 AM UTC
   Should be Finished on: Thu 04 May 2023 12:00:00 AM UTC
                Category: None
                Priority: 5 - Normal
              Item Group: None
                  Status: None
                 Privacy: Public
        Percent Complete: 0%
             Assigned to: None
             Open/Closed: Open
         Discussion Lock: Any
                  Effort: 0.00


    _______________________________________________________

Follow-up Comments:


-------------------------------------------------------
Date: Thu 04 May 2023 08:43:58 PM UTC By: Faezeh Bidjarchian <fbidjarchian>
In machine learning, pooling is a technique to decrease the complexity of the
images (feature maps). 
We consider a set of pixels inside a squared window. The width of this window
is determined by the poolsize.
We apply an operator like maximum (pool-max) on the selected pixels (the
window).
In Gnuastro, there was no function to do this. So, I had to write and add this
program to use in my own job.
I'm delighted to announce that the pooling program is now complete!
You can use any image as an input of this program. It means your input can be
a rectangle or square image.
Also, it can contain blank values.
You must determine the poolsize (that can be any positive and non-zero value
greater than the width of your
image) and the type of the desired operator(poolmax, pool-min, pool-mean,
pool-median).
So if you test this program with the "during-dev.sh" script, you use the
command below:

utilname=arithmetic
arguments="image.fits 2 pool-max"

That image is your desired input, 2 is the size of pooling, and "pool-max" is
the selected operator. 
I tested the program with different images and according to my understanding,
it seems to be working correctly!
If you have any suggestions or comments, or if you encounter any issues while
testing it with other images, please
let me know. 







    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/task/?16362>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/




reply via email to

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