91 replies to this topic

[dciobota]

Thank you for that info. I do notice less darkening at the lower strengths, but then the noise starts showing through. I'll keep experimenting.

[Jasonpitcher]

I really like this application and it had been working great for me. But suddenly, for reasons I havent determined yet, it crashes as soon as I launch it now. Nothing changed that I am aware of my PC so I don't know why it won't run. Reinstalled it but the problem persists. I don't anything helpful in Windows Event Viewer about it, and the app doesn't appear to have any logging of it's own that I've found. 

Anyone run into a similar issue and know a fix by chance? 

[Stanty8]

I've uploaded a new version with some GUI "fixes", really just tried to make things a bit smaller to fit into 768. Hope it helps. 

Uploaded new build "0.3.2-alpha", see first post for link.
 

https://github.com/p...89936a0d0d16e88

Tidy up GUI, separate common areas by color
Add snapshots feature, comparison tool
Add support for native file open/save dialog for platform: win32
Add selector for GPU/CPU
Always save TIF as unit16
Add version to app title

p7ayfu77, I made a fork with the few tweaks I did to get this to work with macOS. Hope that's ok! Allegedly it should work with either Intel or Apple silicon and utilize the GPUs. 

 

Mac users, feel free to try it out and let me know if something doesn't work! 

Edited by Stanty8, 16 February 2023 - 09:31 PM.

[RomainD]

I just downloaded the .EXE and I'm trying different things around but I have absolutely no idea what function is doing what, same for the icons. Is there somewhere a walkthrough the software, some kinf of tutorial or video? So I can understand what I am doing.

For example,

- should you use a FIT linear file or a stretched TIFF file and what to do in those cases?

- What do the slider in the normalize tab?

- What is the impact of the Tiles slider in the noise tab?

- What model is best to use? What is the difference between default and the other?

Edited by RomainD, 23 February 2023 - 05:44 AM.

[Sky King]

Works well, thanks for making this available! I have V0.4.1. Is there a newer one somewhere? Are the new changes just to the GUI? 

[AndreasA]

Hi everyone!

 

So, I came across this topic and I really would love to test this tool on my mac. But I have no idea about computers and I don't understand a single word on this github page nor do I understand how do actually download the program (normally I just need to drag and drop and DMG file to my application folder and thats it...). There is a lot of code and I don't know what I should do if it.

 

Would anybody be so kind and provide me simple to understand explanation how I can install this tool on a mac? I'm very sorry for my Ignorance regarding computers and any help would be highly appreciated!

 

Thanks in advance for any help!

[Stanty8]

Hi everyone!

 

So, I came across this topic and I really would love to test this tool on my mac. But I have no idea about computers and I don't understand a single word on this github page nor do I understand how do actually download the program (normally I just need to drag and drop and DMG file to my application folder and thats it...). There is a lot of code and I don't know what I should do if it.

 

Would anybody be so kind and provide me simple to understand explanation how I can install this tool on a mac? I'm very sorry for my Ignorance regarding computers and any help would be highly appreciated!

 

Thanks in advance for any help!

Hi!

 

So, it took me forever to install on my mac, but I made a few changes and forked the original distribution. Try downloading my version here: https://github.com/s...s/astro-csbdeep , follow the instructions in the readme, especially the parts about installing tensorflow-metal in either conda (if you have an M1 / M2) or venv (if you have an intel chip). If you get errors, let me know.

 

- S

[AndreasA]

Hi!

 

So, it took me forever to install on my mac, but I made a few changes and forked the original distribution. Try downloading my version here: https://github.com/s...s/astro-csbdeep , follow the instructions in the readme, especially the parts about installing tensorflow-metal in either conda (if you have an M1 / M2) or venv (if you have an intel chip). If you get errors, let me know.

 

- S

ok, so I guess one has to be a computer expert for this stuff. There is a lot of code and I don't even know what I should to with...

 

But thank you anyway for your reply and your time!

Clear skies 

[tonyb]

Very nice work. 

 

Posted my testing in the software forum

 

https://www.cloudyni...odenoisepy-041/

 

 

[Stanty8]

ok, so I guess one has to be a computer expert for this stuff. There is a lot of code and I don't even know what I should to with...

 

But thank you anyway for your reply and your time!

Clear skies 

I'm certainly not a computer expert, I promise! All you have to do is install python and the required python packages in the way Apple wants you to (in the readme), download / unzip the code, and run it from your terminal. I'll try to help if any errors pop up for you!

[tonyb]

@jml79:

I followed your steps and I too reduced denoise processing from about 45secs, to 5 secs. I'm on an i7 proc and GTX 1660 GPU.

Note I tried latest CUDA 11.8 - that did not work. 11.2, per your notes below, did work.

 

>>>

Here's what I needed.

I had to install CUDAToolKit 11.2.0 from this link; https://developer.nv...toolkit-archive

Install the CUDA toolkit but select the custom install and deselect all options except the toolkit. If you miss this step the toolkit will downgrade your video driver. This works with the latest 526.98 driver so no need to downgrade it to an older one. It will also warn you about not having MS Visual Studio, select continue anyway.

Next I had to join the NVidia developers network and download cuDNN 8.1.0.77 from here; https://developer.nv...p/cudnn-archive

Copy the contents of the cuDNN zip to the CUDAToolkit folder C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2 It will add the contents of the 3 folders into the existing folders.

Restart AstroDenoisePY

>>>

[jml79]

Yeah, it would be awesome if they just put it in the driver and make CUDA just work transparently. They advertise it as a feature and then you have to bend over backwards and jump through hoops to make it work. The good news is now to make Starnet work all you have to do is one step. It works with 11.2 thankfully. I have both the CLI and GUI version working perfectly. Even integrated with the latest Siril releases. 

[tonyb]

@RomainD, re:

- should you use a FIT linear file or a stretched TIFF file and what to do in those cases?
= I stack and process my images in Siril: calibration, histogram, saturation and save as a 16-bit TIFF. Then I load that TIFF into AstroDenoisePY.

- What do the slider in the normalize tab?
= I leave these alone as I've already some some stretching in Siril.

- What is the impact of the Tiles slider in the noise tab?
= I don't seem to see much impact, so I leave it at the default.

- What model is best to use? What is the difference between default and the other?
= I leave the model at Default.

Hope this helps. A tutorial steps, best workflow practices would be super useful.

[tonyb]

@p7ayfu77 :

 

1. Any additional development work planned on your AstroDenoisePy tool?

2. Any documentation that explains what settings to use?

 

Many thanks for the tool.

 

[RomainD]

@RomainD, re:

- should you use a FIT linear file or a stretched TIFF file and what to do in those cases?
= I stack and process my images in Siril: calibration, histogram, saturation and save as a 16-bit TIFF. Then I load that TIFF into AstroDenoisePY.

- What do the slider in the normalize tab?
= I leave these alone as I've already some some stretching in Siril.

- What is the impact of the Tiles slider in the noise tab?
= I don't seem to see much impact, so I leave it at the default.

- What model is best to use? What is the difference between default and the other?
= I leave the model at Default.

Hope this helps. A tutorial steps, best workflow practices would be super useful.

Many thanks for your answer!

I've been playing around and the tool helped me drastically improve my images, which usually don't have much of integration time (2-3h).

I have found that the default setting reall changed the dynamic of the picture, changing the color balance and blacken the background too much to my liking. Thoug the etting V3.0 something (the other one than default) provide much better natural look to the image and I will use this one from now on.

I've got the best results using a very low value between 100 and 250, just enough to get the details burried in noise stands out.

Would still like some explanation on the Tiling slider if possible.

 

Below are some example comparison before / after the denoising.

 

Left denoise / Right original

 

Left original / Right denoise

 

Left denoise / Right original

Edited by RomainD, 09 March 2023 - 05:13 PM.

[RyzenAMD]

Where can I get a copy of the AstroDenoisePy App?

[tonyb]

@RyzenAMD. https://github.com/p...sbdeep/releases

[AndreasA]

I'm certainly not a computer expert, I promise! All you have to do is install python and the required python packages in the way Apple wants you to (in the readme), download / unzip the code, and run it from your terminal. I'll try to help if any errors pop up for you!

Hi, thank you very much for your help. I really highly appreciate that!

 

So maybe I am completely stupid, but I don't get a single thing from this readme section...

I first went to this link: https://developer.ap...sorflow-plugin/

and then I followed the steps starting with the "x86 : AMD" thing.. I guess that is what I need since I have a old mac with an intel CPU? 

The terminal program did a lot of funny things and then when I came to step 4 "verify" all I got was "command not found" when I entered the first line of this code "import tensorflow as tf"

 

I wanted to just continue with the next steps, but the next point would be "Activate the conda or venv virtual environment created in (2)" - I have no clue how I can activate this venv thing....

Also in the next step "Install the pip3 requirement found in the requirements.txt file by running pip3 install -r requirements.txt" I have no clue how to do that - I guess I would have to download the requirements.txt somehow, but when I click on it, there is just some code and no download option? I am also not sure were I need to save this file (if I would be able to download it)...

 

Sorry for my stupid questions. I really tried hard to google things, but I also don't get the explanations on other pages, as they all use this computer language I really don't get. I am just used to download a dmg file, drag it in my application folder and everything runs fine. I never worked with terminal and this code stuff...

[Astrodroid]

hi 

 

Hi, thank you very much for your help. I really highly appreciate that!

 

So maybe I am completely stupid, but I don't get a single thing from this readme section...

I first went to this link: https://developer.ap...sorflow-plugin/

and then I followed the steps starting with the "x86 : AMD" thing.. I guess that is what I need since I have a old mac with an intel CPU? 

The terminal program did a lot of funny things and then when I came to step 4 "verify" all I got was "command not found" when I entered the first line of this code "import tensorflow as tf"

 

I wanted to just continue with the next steps, but the next point would be "Activate the conda or venv virtual environment created in (2)" - I have no clue how I can activate this venv thing....

Also in the next step "Install the pip3 requirement found in the requirements.txt file by running pip3 install -r requirements.txt" I have no clue how to do that - I guess I would have to download the requirements.txt somehow, but when I click on it, there is just some code and no download option? I am also not sure were I need to save this file (if I would be able to download it)...

 

Sorry for my stupid questions. I really tried hard to google things, but I also don't get the explanations on other pages, as they all use this computer language I really don't get. I am just used to download a dmg file, drag it in my application folder and everything runs fine. I never worked with terminal and this code stuff...

Hi, i just got this working on my Intel Mac. I've done a few things with installing Python based apps in OS X before, and while I consider myself quite technical, I must admit that this method can be quite daunting for the novice, its not so bad if it just works, but when you hit an error or something goes awry, then it can get very confusing.

 

from this URL: https://developer.ap...sorflow-plugin/, under 2. Install base TensorFlow, i had to modify the command or if failed:

(take off the -macos at the end)

 

python -m pip install tensorflow

 

I didn't bother with step 4, and went back to step 3 here https://github.com/s.../main/README.md

 

Step 3, to activate,  run this:

cd /path/to/your/venv (replace with your actual path)

source ./bin/activate

 

Do 4, requirements

Skip 5

6 will run the app - note that the terminal will run in the background while the app is running, so if you do an action in the app, and think nothing is happening, check the terminal. Its rough around the edges in terms of UX, but its a great tool and produced some nice results after a bit of playing around.

Will be using this more, thanks to the devs!!

Edited by Astrodroid, 14 March 2023 - 09:16 AM.

[SamuelWarfel]

from this URL: https://developer.ap...sorflow-plugin/, under 2. Install base TensorFlow, i had to modify the command or if failed:

(take off the -macos at the end)

 

python -m pip install tensorflow

When I try to do this, I get this error: "python -m pip install tensorflow

SyntaxError: invalid syntax", with a little red box over the first p of "pip". 
Any pointers?  I have absolutely no idea what I'm doing, and probably shouldn't be screwing around with this without an expert at my shoulder anyway.
Thanks,
~Sam

[Stanty8]

Hmm. Maybe try omitting python, or trying python3 instead? Depends on your system…

[xploeris]

Followed the instructions for installing and running on Mac. When I try to run it, it fails:

 

File "/Volumes/harddrive/myaccount/venv-metal/lib/python3.10/site-packages/tensorflow/python/framework/load_library.py", line 151, in load_library

py_tf.TF_LoadLibrary(lib)
tensorflow.python.framework.errors_impl.NotFoundError: dlopen(/Volumes/harddrive/myaccount/venv-metal/lib/python3.10/site-packages/tensorflow-plugins/libmetal_plugin.dylib, 0x0006): symbol not found in flat namespace '_TF_GetInputPropertiesList'

Not sure why it's broken or how to proceed.

[Craigar]

Yes I saw some complaints, but there was only second or two to see that console before it shut down...

I had the same problem, here's a successful solution!

I'm running win 10 , no Nvidia card or Cuda - (just embedded Intel video)  and learned from the .log files that all I needed to do was to download and install

 

Visual Studio 2015, 2017, 2019, and 2022

This table lists the latest supported English (en-US) Microsoft Visual C++ Redistributable packages for Visual Studio 2015, 2017, 2019, and 2022....

and selected the "x64.."installer from this link

https://learn.micros...t?view=msvc-170

 

installed , rebooted  and Walla!

my machines info

 

OS Name Microsoft Windows 10 Home
Version 10.0.19045 Build 19045
System Name BlahBlah
System Manufacturer Dell Inc.
System Model XPS 8940
System Type x64-based PC
Processor Intel® Core™ i7-10700 CPU @ 2.90GHz, 2904 Mhz, 8 Core(s), 16 Logical Processor(s)
SMBIOS Version 3.2
Embedded Controller Version 255.255
BIOS Mode UEFI
BaseBoard Manufacturer Dell Inc.
BaseBoard Product 0KV3RP
BaseBoard Version A00
Platform Role Desktop
Secure Boot State On
PCR7 Configuration Elevation Required to View
Windows Directory C:\Windows
System Directory C:\Windows\system32
Boot Device \Device\HarddiskVolume4
Locale United States
Hardware Abstraction Layer Version = "10.0.19041.2728"
User Name 1-122-DELL\1.122
Time Zone Pacific Daylight Time
Installed Physical Memory (RAM) 16.0 GB
Total Physical Memory 15.7 GB
Available Physical Memory 9.38 GB
Total Virtual Memory 18.0 GB
Available Virtual Memory 10.6 GB
Page File Space 2.38 GB
Page File C:\pagefile.sys
Kernel DMA Protection On
Virtualization-based security Not enabled
Device Encryption Support Elevation Required to View
Hyper-V - VM Monitor Mode Extensions Yes
Hyper-V - Second Level Address Translation Extensions Yes
Hyper-V - Virtualization Enabled in Firmware Yes
Hyper-V - Data Execution Protection Yes

 

I learned how to do this from the last line in the log file, which is created every time we start astrodenoisepygui

 

The log files are here

<drive letter>:\Users\<username>\.kivy\logs

here was mine when it crashed, before I did this fix/requirement

 

[INFO   ] Logger: Record log in C:\Users\1.122\.kivy\logs\kivy_23-07-10_2.txt
[INFO   ] Kivy: v2.1.0
[INFO   ] Kivy: Installed at "C:\Users\1.122\AppData\Local\Programs\astrodenoisepygui\lib\kivy\__init__.pyc"
[INFO   ] Python: v3.9.10 (tags/v3.9.10:f2f3f53, Jan 17 2022, 15:14:21) [MSC v.1929 64 bit (AMD64)]
[INFO   ] Python: Interpreter at "C:\Users\1.122\AppData\Local\Programs\astrodenoisepygui\astrodenoisepygui.exe"
[INFO   ] Logger: Purge log fired. Processing...
[INFO   ] Logger: Purge finished!
[INFO   ] Factory: 189 symbols loaded
[INFO   ] Image: Providers: img_tex, img_dds, img_sdl2, img_pil (img_ffpyplayer ignored)
[INFO   ] Window: Provider: sdl2
[INFO   ] GL: Using the "OpenGL" graphics system
[INFO   ] GL: GLEW initialization succeeded
[INFO   ] GL: Backend used <glew>
[INFO   ] GL: OpenGL version <b'4.6.0 - Build 27.20.100.9664'>
[INFO   ] GL: OpenGL vendor <b'Intel'>
[INFO   ] GL: OpenGL renderer <b'Intel® UHD Graphics 630'>
[INFO   ] GL: OpenGL parsed version: 4, 6
[INFO   ] GL: Shading version <b'4.60 - Build 27.20.100.9664'>
[INFO   ] GL: Texture max size <16384>
[INFO   ] GL: Texture max units <32>
[INFO   ] Window: auto add sdl2 input provider
[INFO   ] Window: virtual keyboard not allowed, single mode, not docked
[INFO   ] Text: Provider: sdl2
[WARNING] stderr: Traceback (most recent call last):
[WARNING] stderr:   File "D:\pydeep\astro-csbdeep\.venv\Lib\site-packages\cx_Freeze\initscripts\__startup__.py", line 113, in run
[WARNING] stderr:   File "D:\pydeep\astro-csbdeep\.venv\Lib\site-packages\cx_Freeze\initscripts\Console.py", line 15, in run
[WARNING] stderr:   File "astrodenoisepygui.py", line 29, in <module>
[WARNING] stderr:   File "D:\pydeep\astro-csbdeep\.venv\lib\site-packages\tensorflow\__init__.py", line 41, in <module>
[WARNING] stderr:   File "D:\pydeep\astro-csbdeep\.venv\lib\site-packages\tensorflow\python\__init__.py", line 40, in <module>
[WARNING] stderr:   File "D:\pydeep\astro-csbdeep\.venv\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 28, in <module>
[WARNING] stderr:   File "D:\pydeep\astro-csbdeep\.venv\lib\site-packages\tensorflow\python\platform\self_check.py", line 54, in preload_check
[WARNING] stderr: ImportError: Could not find the DLL(s) 'msvcp140_1.dll'. TensorFlow requires that these DLLs be installed in a directory that is named in your %PATH% environment variable. You may install these DLLs by downloading "Microsoft C++ Redistributable for Visual Studio 2015, 2017 and 2019" for your platform from this URL: https://support.microsoft.com/help/2977003/the-latest-supported-visual-c-downloads

 

 

here's after installing "Microsoft C++ Redistributable for Visual Studio .."

 

 

[INFO   ] Logger: Record log in C:\Users\1.122\.kivy\logs\kivy_23-07-10_3.txt
[INFO   ] Kivy: v2.1.0
[INFO   ] Kivy: Installed at "C:\Users\1.122\AppData\Local\Programs\astrodenoisepygui\lib\kivy\__init__.pyc"
[INFO   ] Python: v3.9.10 (tags/v3.9.10:f2f3f53, Jan 17 2022, 15:14:21) [MSC v.1929 64 bit (AMD64)]
[INFO   ] Python: Interpreter at "C:\Users\1.122\AppData\Local\Programs\astrodenoisepygui\astrodenoisepygui.exe"
[INFO   ] Logger: Purge log fired. Processing...
[INFO   ] Logger: Purge finished!
[INFO   ] Factory: 189 symbols loaded
[INFO   ] Image: Providers: img_tex, img_dds, img_sdl2, img_pil (img_ffpyplayer ignored)
[INFO   ] Window: Provider: sdl2
[INFO   ] GL: Using the "OpenGL" graphics system
[INFO   ] GL: GLEW initialization succeeded
[INFO   ] GL: Backend used <glew>
[INFO   ] GL: OpenGL version <b'4.6.0 - Build 27.20.100.9664'>
[INFO   ] GL: OpenGL vendor <b'Intel'>
[INFO   ] GL: OpenGL renderer <b'Intel® UHD Graphics 630'>
[INFO   ] GL: OpenGL parsed version: 4, 6
[INFO   ] GL: Shading version <b'4.60 - Build 27.20.100.9664'>
[INFO   ] GL: Texture max size <16384>
[INFO   ] GL: Texture max units <32>
[INFO   ] Window: auto add sdl2 input provider
[INFO   ] Window: virtual keyboard not allowed, single mode, not docked
[INFO   ] Text: Provider: sdl2
[WARNING] stderr: AutoGraph is not available in this environment: functions lack code information. This is typical of some environments like the interactive Python shell. See https://github.com/t...-to-source-code for more information.
[INFO   ] GL: NPOT texture support is available
[INFO   ] Base: Start application main loop

 

another very happy camper!

Edited by Craigar, 10 July 2023 - 03:46 AM.

[Tapio]

Good for you!
How long does it take to do its thing?

[Craigar]

the app takes about 45 seconds to fully load

loads 120mb  3704x2776.fits from DSS in about 5 seconds

processes in about 45 seconds using stock params