DeepStack September Update

Hello everyone.
We are happy to share the latest update to DeepStack across all versions including Docker CPU and GPU, Windows CPU and GPU, Jetson and ARM64 with Support for Raspberry PI.

This new update fixes a number of known bugs and errors in the current builds of DeepStack and it provides the most stable windows release yet.
Below are the images:

Docker CPU
deepquestai/deepstack:cpu-2021.09.1

Docker GPU
deepquestai/deepstack:gpu-2021.09.1

Docker Jetson
deepquestai/deepstack:jetpack-2021.09.1

Docker Arm64(Including Raspberry PI)
deepquestai/deepstack:arm64-2021.09.1

Windows CPU
https://github.com/johnolafenwa/DeepStack/releases/download/2021.09.1/DeepStack-Installer-CPU-2021.09.1.exe

Windows GPU
https://github.com/johnolafenwa/DeepStack/releases/download/2021.09.1/DeepStack-Installer-GPU-2021.09.1.exe

Find release notes here; https://github.com/johnolafenwa/DeepStack/releases/tag/2021.09.1

For setup instructions and tutorials, please refer to https://docs.deepstack.cc.

Thanks all for using DeepStack.
We look forward to your feedback always

9 Likes

Thanks!!! I use the Windows CPU. I’ll let you know how it goes.

1 Like

Are we supposed to just install it or uninstall the existing version and delete the folder?

John,

Are you going to update DeepStack to utilize CUDA Toolkit 11.4 and cuDNN v8.2.4 so DeepStack can support GeForce RTX 3000 series GPUs. I was able to manually update the below Windows Packages and have DeepStack work with my RTX 3060

  • numpy-1.21.2-cp37-cp37m-win_amd64
  • Pillow-8.3.2-cp37-cp37m-win_amd64
  • scipy-1.7.1-cp37-cp37m-win_amd64
  • torch-1.9.1+cu111-cp37-cp37m-win_amd64
  • torchvision-0.10.1+cu111-cp37-cp37m-win_amd64

Thanks,
Mike

Are there release notes? What new features, improvements etc. have been made? Thanks!

1 Like

Hello @PaddyDHurley, simply uninstall the previous version and install this

2 Likes

Thanks for the feedback @MikeLud, we plan to provide support for more GPUs soon

2 Likes

Hello @Glen_Cove. This is release is focussed on resolving a number of bugs. We are moving to a montly release cadence for DeepStack and would be publishing release notes soon.

A number of new features and performance improvements is slated for the October Release by end of this month

2 Likes

Great news John, thanks! Love Deepstack and using it with Blue Iris. Can’t wait to see how it matures with further enhancements.

1 Like

Hello, Any chance we could see the updated windows CPU release hosted on the github page? For some reason the download from the provided link seems to be limited to 64Kbs and keeps timing out before the end of the download. Maybe it’s because I’m in the Philippines and there is a region locked bandwidth limit??? Anyways, all the hard work is really appreciated and I can’t wait to try the latest version.

1 Like

I was able to download from the link above using a Tor Browser. There does seem to be a region locked 64Kbps download rate for my region.

1 Like

Hello Nick. We have added release notes and the installers as assets on Github here, https://github.com/johnolafenwa/DeepStack/releases/tag/2021.09.1

1 Like

I’m struggling to pull the pi version with lots of timeouts.

Any tips for how you updated the Windows packages? I think I need to do the same for my 3080. I see them in the DeepStack directory, but not sure how to update. Can it be done with pip or needs to be done differently?

Chris,

In the below link are the steps on how I was able to get DeepStack Windows GPU to work
https://ipcamtalk.com/threads/deepstack-gpu-for-windows-problems-cuda-version-11-and-cudnn-install.59250/post-600376

John,

Any update on the October Release, will it include support for the GeForce RTX 3000 series GPUs.

Thanks
Mike

Is there going to be any clarification on the future hardware acceleration options on Deepstack?

Well aware that deepstack is a 2-3 man effort, and am grateful for what is possible so far.
But it would be helpful though if there was some visibility.