NNStreamer's repositories
nnstreamer
:twisted_rightwards_arrows: Neural Network (NN) Streamer, Stream Processing Paradigm for Neural Network Apps/Devices.
TAOS-CI
:tangerine: TAOS-CI (a.k.a LightSys) is an lightweight and efficient CI system to achieve "Review less, merge faster" with a tool-based review system. It is completely compatible with all GitHub repositories although TAOS-CI is designed & implemented for NNStreamer repository. (Homepage: https://taos-ci.github.io/)
nnstreamer-example
Example applications of nnstreamer. Note that we have to enable the 'apptest" CI module in the near future.
meta-neural-network
Neural Network Support for Yocto / OpenEmbedded
nnstreamer-edge
Remote source nodes for NNStreamer pipelines without GStreamer dependencies
nnstreamer-ros
NNStreamer extension plugins for ROS support
nnstreamer-android-resource
NNStreamer Android Studio Support Resource Files
deviceMLOps.MLAgent
TBD: deviceMLOps.service or deviceMLOps.MLAgent.
reference-models
Neural network models for reference boards (including NPU)
nnstreamer-extra
Experimental features that require additional external dependencies.
deviceMLOps.learning.framework
Learning Framework manages neural network training tasks on devices, which can be a service handler for device MLOps's service daemon or can be an independent application service.
deviceMLOps.service
The main service daemon of Device MLOps. Its interface is defined at api.git as ML.Service and ML.Service.Extension.
run-on-arch-action
A Github Action that executes jobs/commands on non-x86 cpu architectures (ARMv6, ARMv7, aarch64, s390x, ppc64le) via QEMU
vivante-npu-lib
[TODO] This repository is to generate a low-level library package from the Vivante NPU toolkit.