deepstack in dire straights?

[githubDiversity]

I have reason to believe the project is dying. I hope I am mistaken.

Given that there currently is no known reliable/working deepstack windows GPU version that I am aware of I decided to get the code and do it myself.

What I found was a surprise to me and let me just say that things need to be cleaned up real fast in order to have any life expectancy.

I have reached out to @john on github regarding the state the current code is in but havn’t heard back yet.

If there is anyone with a working windows GPU version (preferably a recent one) please zip it up and share it. I can decompile the code if need be and start from there in an effort of cleaning house.

[jgosnell56]

This is the latest version I have. https://www.filehosting.org/file/details/956874/DeepStack-Installer-GPU-2021.02.1.zip

[githubDiversity]

thank you for your efforts @jgosnell56 I have downloaded it just to be sure. Baswed on the filename and filesize this is the version I also tried. What I meant was if someone can please zip up their installed C:\Deepstack folder. If that does work on my end I can then try to get the most recent source code to produce an output similar to that (only cleaned up and with the most recent code)

[jgosnell56]

I can do that as well. Give me just a bit.

[jgosnell56]

Try this. https://www.filehosting.org/file/details/957003/DeepStack.zip

[githubDiversity]

thanks a bunch, I'll keep one updated on how far I get.

[githubDiversity]

finally I have zeroed it down.

I can now try to compile a working 2021-6-1 windows GPU version. Are you interested and if so why? It' will help me motivate to push through

[jgosnell56]

I would certainly be interested in it. I run deep stack on a home server with a decent NVIDIA graphics card. I have found that the GPU version of DeepStack as far less latency than the CPU version. I assume there are many others in the same boat or they have the same feelings.

On Wed, Aug 11, 2021, 1:00 PM githubDiversity @.***> wrote:

finally I have zeroed it down.

I can now try to compile a working 2021-6-1 windows GPU version. Are you interested and if so why? It' will help me motivate to push through

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/VorlonCD/bi-aidetection/issues/269#issuecomment-897033973, or unsubscribe https://github.com/notifications/unsubscribe-auth/AQDXQUTI6W56DLOZFVOXWMDT4K3EBANCNFSM5BWC4IAA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email .

[githubDiversity]

ugg. I am running into another wall when running a compiled windows GPU based on the latest available source code. TL;DR; I have a lot more digging and error resolving to to before I can even begin of cleaning house. I will try but I assume the developers abandoned, if this is actually the case, it for a good reason which up until now I do not know yet.

For the tech savvy amongst us; Process Process-2: Traceback (most recent call last): File "D:\obj\windows-release\37amd64_Release\msi_python\zip_amd64\process.py", line 297, in _bootstrap File "D:\obj\windows-release\37amd64_Release\msi_python\zip_amd64\process.py", line 99, in run File "C:\DeepStack\intelligencelayer\shared\face.py", line 73, in face cuda=SharedOptions.CUDA_MODE, File "C:\DeepStack\intelligencelayer\shared.\recognition\process.py", line 28, in init load_model(self.model, model_path) File "C:\DeepStack\intelligencelayer\shared.\recognition\process.py", line 11, in load_model checkpoint = torch.load(path, map_location=lambda storage, loc: storage) File "C://DeepStack\windows_packages\torch\serialization.py", line 585, in load return _legacy_load(opened_file, map_location, pickle_module, pickle_load_args) File "C://DeepStack\windows_packages\torch\serialization.py", line 755, in _legacy_load magic_number = pickle_module.load(f, pickle_load_args) _pickle.UnpicklingError: invalid load key, 'v'.

[Tinbum1]

I've found the developers seem to work in spurts- they go quiet and then they suddenly reappear.

[bigshow104]

Has there been any progress on this. I am using aitool and can not get it toi work with deep stack. I did have it running on a docker but am now running hte windows gpu version but still not change.

[githubDiversity]

It seems there is no more need as the developer of deepstack said that a compiled windows GPU version will now be included in the release cycles. That was good news as I got intimidated by the code soup that I perhaps could not even get through if I kept at it relentlessly