https://askubuntu.com/questions/1228423/how-do-i-fix-cuda-breaking-after-suspend
This is great! Now, I can use CUDA after suspend. #ubuntu, #linux, #deepstream
https://askubuntu.com/questions/1228423/how-do-i-fix-cuda-breaking-after-suspend
This is great! Now, I can use CUDA after suspend. #ubuntu, #linux, #deepstream
Guess I shouldn’t have tempted fate with a post like that. It failed again after several hours of running.
Trying another idea now…
#gstreamer #nvidia #software #softwareDevelopment #ml #MachineLearning #ComputerVision #DeepStream
I think (hope) I have finally won my battle with #gstreamer and #nvidia and we have worked around the insidious memory leak (that's on Nvidia's nvvideoconvert component we believe, which seems to leak when a pipeline containing it is destroyed), and the frustrating failure of the GStreamer qtdemux to emit its pad creation signal intermittently (very intermittently).
#software #softwareDevelopment #ml #MachineLearning #ComputerVision #DeepStream
Upgrading the "slim" base container for our #Deepstream app would be a lot simpler if #Nvidia didn't keep restructuring how they package these things.
All they need to do is package this stuff in clean Debian packages and it would be trivial to upgrade, but no, instead they use a mix of packages (including packages that unpack a local repo they then install packages from!), tar balls, scripts and then they move & delete files randomly too. It's a mess.
Как просто добавить ИИ в приложения на Rust: универсальный опенсорсный инструмент
Системный разработчик ИТ-компании «Криптонит» написал статью про новый инструмент на Rust, который облегчает запуск моделей машинного обучения и их внедрение в приложения. Дальше публикуем текст от первого лица. Статья написана по материалам выступления Михаила на RustCon 2024. Посмотреть видеозапись доклада можно в VK Видео .
https://habr.com/ru/companies/kryptonite/articles/873286/
#rust #библиотека #машинное_обучение #ml #модели #triton #deepstream #tensorrt #cuda #ии
"nvds_batch_meta_release_func()
batch_meta release function called when meta_data is going to be released."
When the documentation simply rephrases the name of the function, it does not add much value. #Nvidia #Deepstream
Now we have @dmorin talking about analytics and inference in #GStreamer.
In addition to doing media processing, GStreamer also allows carrying metadata with media, as well as representing data pipelines that have no media at all.
GStreamer / Deepstream have to be the most frustrating things out there. Every change makes it fail in new ways, none of which make any sense, but none work the way they should.