Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

StarNet v2

Astrophotography Imaging Software
  • Please log in to reply
421 replies to this topic

#51 nekitmm

nekitmm

    Vendor - StarNet Software

  • *****
  • Vendors
  • topic starter
  • Posts: 227
  • Joined: 25 Apr 2018

Posted 29 January 2022 - 01:19 AM

Very nice to see this new release of Starnet version 2, Nikita, thank you!  I was happy to find that to continue successfully using the CUDA-boosted performance of Starnet++, all I had to do was copy the tensorflow.dll file that I used in Starnet++ over to the StarnetGUI_Win folder.

As others, I too would like to see a command-line version of Starnet version 2 as you provided in the original Starnet.  In contrast to the simple GUI you created, I favor the approach I took for Starnet++  for which I created a batch script to invoke Starnet++ from the Windows Context Menu "Send To" command.  That allowed me to very conveniently right-click on the file I wanted to process, select Send To Starnet++, and receive prompting for the stride size, with the output file then automatically renamed and written back to the same folder as the original file.  I did attempt that with starnetGUI.exe file, but quickly saw it ignored the arguments provided by the script and just ran in the usual manner.

Here's my own comparison image...nice results!  https://tinyurl.com/dp-Starnet1vs2
Thanks!

Daryl 

Wow, the command-line version is more popular than I imagined, lol! I will work on it when I got time!

 

The improvement on your image is huge, glad it works for you!


  • Ambress likes this

#52 nekitmm

nekitmm

    Vendor - StarNet Software

  • *****
  • Vendors
  • topic starter
  • Posts: 227
  • Joined: 25 Apr 2018

Posted 29 January 2022 - 01:20 AM

Is there any way to use this on Mac without having PI? Can I use it as a standalone product rather than a PI module, since I don't have PI.

Thanks

As I mentioned, there could be a way to run Windows GUI version on Macs, but I will also work on command-line version!


  • ktran likes this

#53 nekitmm

nekitmm

    Vendor - StarNet Software

  • *****
  • Vendors
  • topic starter
  • Posts: 227
  • Joined: 25 Apr 2018

Posted 29 January 2022 - 01:22 AM

Just installed and the first impression is that the results are incredible! This may force me to go back and reprocess a bunch of old data.

 

The instructions say to put everything in PI's bin directory but the weights file needs to go in PI's lib directory.

 

And regarding installation, any chance that a PI repository could be created for this? It would simplify installation and make it easy for people to stay up to date.

I think it will be updated as soon as PI team has time. They are currently busy with more crucial updates!



#54 nekitmm

nekitmm

    Vendor - StarNet Software

  • *****
  • Vendors
  • topic starter
  • Posts: 227
  • Joined: 25 Apr 2018

Posted 29 January 2022 - 01:24 AM

i read all of his points to mean that you can't do any of those things in support of someone trying to develop another neural-net based star removal program. of course the reason we have lawyers is that human languages can be ambiguous, so this is not a problem unique to SN2's T&C. hopefully nikita can clarify.

 

rob

Yeah, this is correct interpretation, Rob. I thought the message was pretty clear, definitely not going to impose crazy restrictions here...



#55 JeffreyHorne

JeffreyHorne

    Vostok 1

  • -----
  • Posts: 117
  • Joined: 22 Apr 2013

Posted 29 January 2022 - 02:28 AM

@nekitmm, this is working so well! The best star removal (and star mask) that I've seen.

 

I'm using it with my 3080 and it just flies through, even on an 2x upsampled image. (Edit: when I try linear 2x upsampled with a starmask generated, it gets to 100% complete and then Pixinsight crashes hard, closes itself. Not a big deal, as it works great on stretched images.)

 

A couple of questions:

 

1. Getting it to work on my 3080 was as simple as others stated...just leave the original tensorflow file in place that was used for Starnet V1 CUDA. Is there any drawback to doing this? Does it matter that it's not the same file (or possibly not the same version) of the tensorflow file that's packaged with Starnet V2?

 

2. Is there any downside to upsampling 2x, other than increased processing time? Processing time is so quick with a good graphics card that I don't mind waiting longer, and I want to make sure that I'm getting the best quality that I can from the app.

 

Thanks again for this gift!

Jeff


Edited by JeffreyHorne, 29 January 2022 - 03:35 AM.

  • nekitmm likes this

#56 kaicyung

kaicyung

    Explorer 1

  • *****
  • Posts: 63
  • Joined: 27 Jul 2016

Posted 29 January 2022 - 04:03 AM

Rob,

 

I honesty didn't specify any custom compilation flags when compiling TF... I guess if it does work no your M1 Mac, then it is the default...

 

The newer StarNet version does use newer TF API, so you have to use new TF library to run it. However, it should be backwards-compatible. Weird that on my Mac both versions of StarNet work with the libraries provided with newer StarNet! I wonder why that's not your case...

First, fantastic news. Can't wait to try v2 out. 

 

Question, what tensorflow and cuda version did you use to train your neural network? I can help work out the steps to use the newer cuda support files for the AP community.

 

Kai



#57 _MiR_

_MiR_

    Lift Off

  • -----
  • Posts: 8
  • Joined: 28 Jan 2022

Posted 29 January 2022 - 05:30 AM

Thanks to pfile's post I was able to install StarNet++ on both my M1 Mac and my intel mac.

 

Here is a summary of the steps to do from a terminal session:

 

Download Starnet from SourceForge, make sure the zip file is in your Downloads directory

open terminal window

We need to copy files as root, so when asked enter your Password so that we can elevate user rights to be able to copy everything

 

M1:

cd $HOME/Downloads
unzip -o StarNetv2PI_MacOS.zip
cd StarNetv2PI_MacOS

sudo cp lib/StarNet2_weights.pb /Applications/PixInsight/lib/
sudo chown root:admin           /Applications/PixInsight/lib/StarNet2_weights.pb
sudo chmod 644                  /Applications/PixInsight/lib/StarNet2_weights.pb

sudo cp StarNet2-pxm.dylib /Applications/PixInsight/bin/
sudo chown root:admin      /Applications/PixInsight/bin/StarNet2-pxm.dylib
sudo chmod 755             /Applications/PixInsight/bin/StarNet2-pxm.dylib

sudo rm -f             /Applications/PixInsight/bin/libtensorflow*
sudo cp libtensorflow* /Applications/PixInsight/PixInsight.app/Contents/Frameworks/
sudo chown root:admin  /Applications/PixInsight/PixInsight.app/Contents/Frameworks/libtensorflow*
sudo chmod 755         /Applications/PixInsight/PixInsight.app/Contents/Frameworks/libtensorflow*

intel:

PixInsight does come with tensorflow libs version 2.6.0 and they seem to work for the new Starnetv2, so we do not need to copy the libtensorflow libs to pixinsight. 

cd $HOME/Downloads
unzip -o StarNetv2PI_MacOS.zip
cd StarNetv2PI_MacOS

sudo cp lib/StarNet2_weights.pb /Applications/PixInsight/lib/
sudo chown root:admin           /Applications/PixInsight/lib/StarNet2_weights.pb
sudo chmod 644                  /Applications/PixInsight/lib/StarNet2_weights.pb

sudo cp StarNet2-pxm.dylib /Applications/PixInsight/bin/
sudo chown root:admin      /Applications/PixInsight/bin/StarNet2-pxm.dylib
sudo chmod 755             /Applications/PixInsight/bin/StarNet2-pxm.dylib

in case your cats explodes or the pillars of creation suddenly show up in all your andromeda pics you can go the whole nine yards and also copy over the tensorflow libs on intel:

cd $HOME/Downloads
unzip -o StarNetv2PI_MacOS.zip
cd StarNetv2PI_MacOS

sudo cp lib/StarNet2_weights.pb /Applications/PixInsight/lib/
sudo chown root:admin           /Applications/PixInsight/lib/StarNet2_weights.pb
sudo chmod 644                  /Applications/PixInsight/lib/StarNet2_weights.pb

sudo cp StarNet2-pxm.dylib /Applications/PixInsight/bin/
sudo chown root:admin      /Applications/PixInsight/bin/StarNet2-pxm.dylib
sudo chmod 755             /Applications/PixInsight/bin/StarNet2-pxm.dylib

sudo rm -f             /Applications/PixInsight/bin/libtensorflow*
sudo cp libtensorflow* /Applications/PixInsight/bin/
sudo chown root:admin  /Applications/PixInsight/bin/libtensorflow*
sudo chmod 755         /Applications/PixInsight/bin/libtensorflow*

In an earlier post mrszymon mentioned that it may be necessary to change file attributes, I did not need to do that but here are the lines again for the sake of completeness:

sudo xattr -dr com.apple.quarantine /Applications/PixInsight/bin/StarNet2-pxm.dylib

After those steps follow the usual steps to register Starnet2 in pixinsight.


  • rawhead909, airscottdenning, pfile and 9 others like this

#58 Roger123456

Roger123456

    Lift Off

  • -----
  • Posts: 10
  • Joined: 07 Oct 2020

Posted 29 January 2022 - 07:50 AM

I'm looking forward to getting this working, but currently getting a " Could not find checkpoint file in 'C:\Program Files\Pixinsight\bin\'! "  error.   I've tried all the obvious things such as redownloading StarNet v2.  I believe that I have all the files in the right places (details below).  I have tried the old fix (no longer needed) of changing directory in the Process console to point to the Pixinsight bin directory.  I have the most up to date version of PI installed.  The original Starnet still works ok.

Before uninstalling PI and reinstalling, I'd like to see if someone on here can see what I'm doing wrong...  
 

 Windows 10
 PixInsight 1.8.8-12 (x64)
 "No updates currently available"

  Downloaded the Windows PI version of Starnet v2

  Put StarNet2-pxm.dll and tensorflow.dll into Program Files/PixInsight/bin   (overwrote the previous version of tensorflow.dll, tried using old version too, same)
  Put Starnet2_weights.pb into Program Files/PixInsight/library (along with the old mono_ and rgb_starnet_weights) .    I also tried putting this in Program Files/PixInsight/lib and Program Files/PixInsight/lib/x64.     [I suspect that this multitude of options of where to put it has something to do with it]

  Opened PI

   Run Starnet2 stride 256 with create starmask ticked.

On Process Console:

Starting star removal procedure...

Window size: 512
Stride: 256
Image size: 6209x4141
Number of channels: 3
Color space: RGB
Bits per sample: 32
Has alpha channels: false
Float sample: true
Restoring neural network checkpoint...
*** Error: Could not find checkpoint file in 'C:\Program Files\PixInsight\bin\'!
Reading swap files...
4515.056 MiB/s
<* failed *>


Any ideas?


  • Tim C likes this

#59 dghent

dghent

    N.I.N.A. (& More)

  • *****
  • Freeware Developers
  • Posts: 1,958
  • Joined: 10 Jun 2007
  • Loc: Maryland, US

Posted 29 January 2022 - 10:05 AM

Yeah, I guess I need to add 'software' to this phrase to have 'any derivative software'. But overall you just pulled this phrase out of the context and opened it up to your free interpretations, which clearly contradict what is stated above: updated license terms that specifically limit StarNet usage to astrophotography processing only. So the output can obviously be processed. I am not sure why I have to write this.

 

Again, pulled out of context, just read the previous sentence.

 

Clearly states that StarNet shall not be used to aid development of other software. I don't see how StarNet is used to develop EZ Suite. EZ Suite merely uses StarNet the same way any end user can. In other words, it just incorporates StarNet into it's workflow. That's totally fine.

 

I explained a bit how neural networks-based software is different from traditional software in my post. This includes the explanation of how this can be harmful. I agree that this is one of more 'easy to misinterpret' pieces, but again this phrase should be considered in the context of aiding the development of derivative software.

 

This license most certainly does not prevent anyone from using StarNet for any meaningful purpose. Furthermore, even if you pay for the product it does not necessarily give you the privilege to do whatever you wish with the program's output, lol. Again, I explained why I am imposing some limitations on the usage in my post. I can not sell it or even receive donations for it at the moment.

 

 

Well, sorry, but things are just a bit more complicated in the case of neural networks, that's not my fault.

 

 

I am always available to clarify the license terms if needed, but I am not going to play the game of 'pull a phrase out of the context and come up with crazy interpretations'. I think the overall message is pretty clear and the license is not restrictive at all if your intent is to create astro images.

 

Hi, and thank you for responding. I first want to give some background on myself so you can see where I come from on this topic. I've been the maintainer of one free open source operating system and currently am on a team that develops and maintains a custom, mixed open source operating system for storage systems. Given that OSes are large amalgamations of a wide variety of open source, closed source, and commercial software components, we have to be on the ball regarding the wide variety of licenses that are involved in maintaining OSes. So, while I'm no lawyer, being able to read licenses and ensure that they are being accommodated and are compatible with how we wish to employ a given piece of software is a necessary ability of my daily job.

 

I believe the issue here is that you're leaning far too heavily on implied context in your license stipulations. Licenses are meant to be standalone documents that lay out each term in absolute, concise, and non-subjective language. The terms cannot silently associate themselves with some larger, unseen context and expect the reader to properly realize that connection. Doing that in a "but you know what I mean" way or expecting the reader to find this CN thread first to - maybe - realize the context introduces ambiguity, and ambiguity is what will make people either ignore the license entirely or (in my case, as a software developer) avoid the software it's connected to altogether. The obliviousness to context can be especially true when it comes to non-English speaking people and how they read and understand the terms directly.

 

So, when you say "StarNet shall not be used to aid development of other software."(period bolded for emphasis), that statement must be read and parsed standing on its own without the requirement that the reader already be aware of some special context. By itself and without any stated qualifiers as to what "development" constitutes or what "other software" actually is, it will simply read as StarNet2 cannot be used in any kind of software development in any capacity. So, if I'm developing software and want to use StarNet2 as a mask generation tool, I cannot use StarNet2 because doing so would be aiding the development of my own software, which would be a violation of that license term.

 

If your counter to that assessment begins with "Yes, but..." or "That's not what I meant...", then that is a clear indication that the license terms are being inadequately stated. It's the same deal when you generically refer to "outputs" being used to create "derivative works" or when sharing "output" with other people (by the way, what does "works well" really mean, anyway?) You do have to be explicit in how you state these terms and list any exceptions or specific allowances along with them so that the user is fully aware of what's permissible and what's not. Yeah it's pain, but this kind of rigor is what one signs up for when restrictions on use want to be conveyed.

 

I don't want to belabor this but I really do believe that the terms need to made clear on their own and I hope you're able to find the time to do that. It would be tragic if the software wasn't utilized in what would be an allowed manner due to an ambiguity or misinterpretation of the terms. In fact, this happening with a colleague who maintains an automated PixInsight processing pipeline is what spurred me to post. He read the StarNet2 announcement and was highly confused by how it will affect the development of his own work and if he would be "on the hook" for ensuring that users of his PI scripts also adhere to the terms regarding how "output" is used. He's not a CN user, so I asked on his behalf. I didn't bring this topic up because I wanted to argue about something :)


  • dswtan, Ruediger, polslinux and 2 others like this

#60 _MiR_

_MiR_

    Lift Off

  • -----
  • Posts: 8
  • Joined: 28 Jan 2022

Posted 29 January 2022 - 10:19 AM

I did a little more testing on performance on M1 vs. Intel Mac.

To do so I downloaded official tensorflow build from google, used the provided libs from inside of Starnet and just for fun compiled tensorflow 2.9 with avx and avx2 support.

Tests were done on a Macbook Pro 16" 2.3 GHz 8-Core Intel Core i9 and a Macbook Air M1 (2020)

 

I took the same greyscale image and processed it three times with different tensorflow libs:

 

Window size: 512
Stride: 256
Image size: 6224x4168
Number of channels: 1
Color space: Grayscale
Bits per sample: 32
Has alpha channels: false
Float sample: true

 

Official tensorflow v 2.7.0 (seems to have avx2 anabled):

Intel: 01:54 Minutes processing time

M1: Crashes with unsupported OpCode

 

Starnet provided lib:

Intel: 01:42 Minutes

M1: 04:36

 

DiY Build tensorflow v 2.9.0 with avx2 enabled:
Intel: 01:47

M1: Crashes with unsupported OpCode

 

I am currently building tensorflow 2.9.0 without avx2 (and perhaps also other versions to see if by chance I find a faster version for the M1.

 

I have no idea why the non-avx2 version that comes with starnet is roughly as fast (in fact faster) as the other libs with avx2.

 

Commandline used to build avx2 version of tensorflow:

 

bazel build -c opt --copt=-msse4.1 --copt=-msse4.2 --copt=-mavx --copt=-mavx2 --copt=-mfma --copt=-O3 //tensorflow/tools/lib_package:libtensorflow


  • pfile and nekitmm like this

#61 RT--

RT--

    Lift Off

  • -----
  • Posts: 20
  • Joined: 12 Aug 2017

Posted 29 January 2022 - 10:24 AM

I'm looking forward to getting this working, but currently getting a " Could not find checkpoint file in 'C:\Program Files\Pixinsight\bin\'! "  error.   I've tried all the obvious things such as redownloading StarNet v2.  I believe that I have all the files in the right places (details below). 
...

Put Starnet2_weights.pb into Program Files/PixInsight/library 

Weights also need to be in bin. This is also stated in the readme file along with the error message in the troubleshooting section.


  • nekitmm likes this

#62 ckoval7

ckoval7

    Lift Off

  • -----
  • Posts: 1
  • Joined: 29 Jan 2022
  • Loc: Pasadena, MD

Posted 29 January 2022 - 10:42 AM

I'm looking forward to the Linux version.


  • damianc likes this

#63 daviswh

daviswh

    Sputnik

  • -----
  • Posts: 26
  • Joined: 16 Oct 2021

Posted 29 January 2022 - 11:16 AM

I really hate asking stupid questions, but how to I install the stand alone version on a Windows laptop? I downloaded the StarNetv2GUI_Win.zip then extracted the zip file. As far as I can tell there is no executable file. 

Again - sorry for being stupid.

Thanks

Bill



#64 Roger123456

Roger123456

    Lift Off

  • -----
  • Posts: 10
  • Joined: 07 Oct 2020

Posted 29 January 2022 - 11:21 AM

Weights also need to be in bin. This is also stated in the readme file along with the error message in the troubleshooting section.

Thanks!, that fixed it after I disabled and then re-installed the Starnet2 module.


  • nekitmm likes this

#65 afryhover

afryhover

    Sputnik

  • -----
  • Posts: 44
  • Joined: 17 Aug 2012
  • Loc: Yukon, Oklahoma

Posted 29 January 2022 - 02:35 PM

Thank you very much for all your hard work and efforts!


  • nekitmm likes this

#66 Ambress

Ambress

    Lift Off

  • -----
  • Posts: 18
  • Joined: 14 Apr 2021

Posted 29 January 2022 - 04:59 PM

I really hate asking stupid questions, but how to I install the stand alone version on a Windows laptop? I downloaded the StarNetv2GUI_Win.zip then extracted the zip file. As far as I can tell there is no executable file. 

Again - sorry for being stupid.

Thanks

Bill

Bill, it sounds like you downloaded the correct zip file, so you should in fact find that there is an "exe" file, unless you've got Windows hiding the file exensions.  Regardless, you will want to run the file named "starnetGUI" (.exe unless hidden) and it requires no installation, running as a stand-alone executable from within the extracted folder. 


Edited by Ambress, 29 January 2022 - 04:59 PM.


#67 pfile

pfile

    Fly Me to the Moon

  • -----
  • Posts: 6,402
  • Joined: 14 Jun 2009

Posted 29 January 2022 - 06:23 PM

I did a little more testing on performance on M1 vs. Intel Mac.

To do so I downloaded official tensorflow build from google, used the provided libs from inside of Starnet and just for fun compiled tensorflow 2.9 with avx and avx2 support.

Tests were done on a Macbook Pro 16" 2.3 GHz 8-Core Intel Core i9 and a Macbook Air M1 (2020)

 

I took the same greyscale image and processed it three times with different tensorflow libs:

 

Window size: 512
Stride: 256
Image size: 6224x4168
Number of channels: 1
Color space: Grayscale
Bits per sample: 32
Has alpha channels: false
Float sample: true

 

Official tensorflow v 2.7.0 (seems to have avx2 anabled):

Intel: 01:54 Minutes processing time

M1: Crashes with unsupported OpCode

 

Starnet provided lib:

Intel: 01:42 Minutes

M1: 04:36

 

DiY Build tensorflow v 2.9.0 with avx2 enabled:
Intel: 01:47

M1: Crashes with unsupported OpCode

 

I am currently building tensorflow 2.9.0 without avx2 (and perhaps also other versions to see if by chance I find a faster version for the M1.

 

I have no idea why the non-avx2 version that comes with starnet is roughly as fast (in fact faster) as the other libs with avx2.

 

Commandline used to build avx2 version of tensorflow:

 

bazel build -c opt --copt=-msse4.1 --copt=-msse4.2 --copt=-mavx --copt=-mavx2 --copt=-mfma --copt=-O3 //tensorflow/tools/lib_package:libtensorflow

interesting performance numbers. seems like avx2 might not matter on this particular graph. i have noticed that my CPU utilization is never anywhere near 100% when running starnet on my i9-9960x which has 32 threads. maybe this workload just isn't parallelizable enough for avx vs. non-avx to matter.

 

rob



#68 linuxkidd

linuxkidd

    Explorer 1

  • *****
  • Posts: 52
  • Joined: 05 Feb 2012
  • Loc: Las Vegas, NV

Posted 29 January 2022 - 07:15 PM

Hi all,

  Wondering if there's any tips for high resolution image processing... I've never had very good luck with pre 2.0 versions of starnet, but the 2.0 looks quite a bit better.  Alas, I'm still not getting the perfectly starless results I see others posting.  I may just not be setting things up properly for starnet to succeed, so looking for any assistance.

 

Camera: QHY600, 1x1 binning ( 61 megapixels )

Scope: Planewave CDK 12.5 (yes, it's way over sampled at ~0.3arc-sec per pixel)

Target: Lower's Nebula SH2-297

 

Technique:

 - Register, Align, Integrate each channel

 - Mure Denoise the individual channels

 - Channel combine

 - Photometric Color Calibration

 - Histogram stretch

 - Starnet run

 

Original non-linear image:
Lowers-Nebula-screenshot4.png

 

Starnet ( provided with PI )
Lowers-Nebula-starnetv1.png

 

 

Starnetv2:
Lowers-Nebula-starnetv2.png

 

Thanks in advance!  ( and sorry for the large photos... figured the larger size would help identify the issue better?

Michael

aka linuxkidd


  • danny1976 likes this

#69 linuxkidd

linuxkidd

    Explorer 1

  • *****
  • Posts: 52
  • Joined: 05 Feb 2012
  • Loc: Las Vegas, NV

Posted 29 January 2022 - 07:43 PM

OK, a bit more success... this time, when I stretched the image, I smashed the blacks quite a bit.

 

Lowers-Nebula-starnetv2-smashed-blacks.p



#70 nekitmm

nekitmm

    Vendor - StarNet Software

  • *****
  • Vendors
  • topic starter
  • Posts: 227
  • Joined: 25 Apr 2018

Posted 29 January 2022 - 09:20 PM

OK, a bit more success... this time, when I stretched the image, I smashed the blacks quite a bit.

Overall the problem is that your images are undersampled I think - with 61 megapixels those stars are probably huge and therefore not processed too well. You can try and downsample it for experiment - the result should be much better...

 

I can also tell that your kind of data (high FL and high SNR) is the data StarNet performs the worst on...


Edited by nekitmm, 29 January 2022 - 09:23 PM.

  • linuxkidd likes this

#71 nekitmm

nekitmm

    Vendor - StarNet Software

  • *****
  • Vendors
  • topic starter
  • Posts: 227
  • Joined: 25 Apr 2018

Posted 29 January 2022 - 09:22 PM

I really hate asking stupid questions, but how to I install the stand alone version on a Windows laptop? I downloaded the StarNetv2GUI_Win.zip then extracted the zip file. As far as I can tell there is no executable file. 

Again - sorry for being stupid.

Thanks

Bill

That's weird, the exe file should be there! Can you upload the list of files you see?



#72 nekitmm

nekitmm

    Vendor - StarNet Software

  • *****
  • Vendors
  • topic starter
  • Posts: 227
  • Joined: 25 Apr 2018

Posted 29 January 2022 - 09:43 PM

I believe the issue here is that you're leaning far too heavily on implied context in your license stipulations. Licenses are meant to be standalone documents that lay out each term in absolute, concise, and non-subjective language. The terms cannot silently associate themselves with some larger, unseen context and expect the reader to properly realize that connection. Doing that in a "but you know what I mean" way or expecting the reader to find this CN thread first to - maybe - realize the context introduces ambiguity, and ambiguity is what will make people either ignore the license entirely or (in my case, as a software developer) avoid the software it's connected to altogether. The obliviousness to context can be especially true when it comes to non-English speaking people and how they read and understand the terms directly.

I am not relying on any implied (or unseen) context - I am talking about what is written in the license, e.g. previous sentence. Is that too much to ask?

 

 

I don't want to belabor this but I really do believe that the terms need to made clear on their own and I hope you're able to find the time to do that. It would be tragic if the software wasn't utilized in what would be an allowed manner due to an ambiguity or misinterpretation of the terms. In fact, this happening with a colleague who maintains an automated PixInsight processing pipeline is what spurred me to post. He read the StarNet2 announcement and was highly confused by how it will affect the development of his own work and if he would be "on the hook" for ensuring that users of his PI scripts also adhere to the terms regarding how "output" is used. He's not a CN user, so I asked on his behalf. I didn't bring this topic up because I wanted to argue about something smile.gif

I still don't fully agree with your points, but now I at least see what is the problem. This is something specific enough to discuss. I will shoot you a PM so we can continue this discussion in more specific and productive manner. Before I read this, I had no idea where you were going to with your posts to be honest, it did seem to me like an argument just for the sake of argument.


Edited by nekitmm, 29 January 2022 - 11:38 PM.

  • dswtan and CptNautilus like this

#73 daviswh

daviswh

    Sputnik

  • -----
  • Posts: 26
  • Joined: 16 Oct 2021

Posted 30 January 2022 - 12:08 PM

Bill, it sounds like you downloaded the correct zip file, so you should in fact find that there is an "exe" file, unless you've got Windows hiding the file exensions.  Regardless, you will want to run the file named "starnetGUI" (.exe unless hidden) and it requires no installation, running as a stand-alone executable from within the extracted folder. 

Ambress,

I was able to run the program. When I ran the starnetGUI I received a warning that MSVCP140.dll was missing. I installed the .dll and now it works fine. I'd prefer to run Starnett standalone on my iMAC, it has a lot more horsepower than my laptop. I bought a cheap laptop just to run DSS, now its morphing...

 

Thanks for the help

Bill



#74 JeffreyHorne

JeffreyHorne

    Vostok 1

  • -----
  • Posts: 117
  • Joined: 22 Apr 2013

Posted 30 January 2022 - 04:59 PM

Anybody else have issues with Pixinsight crashing after Starnet V2 completes? No error message, Pixinsight just shuts down. I’m using 2x upsampling and 128 stride, with GPU acceleration from a 3080. I kept the original tensorflow file from Starnet 1 to get GPU acceleration working, and I’m wondering if that’s causing my crashes.  



#75 T~Stew

T~Stew

    Vanguard

  • -----
  • Posts: 2,002
  • Joined: 29 Sep 2020
  • Loc: central Ohio

Posted 30 January 2022 - 10:37 PM

Wow, the command-line version is more popular than I imagined, lol! I will work on it when I got time!

 

The improvement on your image is huge, glad it works for you!

Looking forward to Linux (and hopefully same simple command line interface) here!

 

Command line was easy, since I always name my file the same thing, just a mater of right click to open terminal and hit the up arrow to go to last command, and hit enter. Couldn't really make it much easier.


  • nekitmm likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics





Also tagged with one or more of these keywords: Astrophotography, Imaging, Software



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics