This layer provides support for building Machine learning and AI applications and related software.

Git repository

https://github.com/Freescale/meta-freescale-ml web repo

Last commit: 2 years, 2 months ago (kirkstone branch)

Maintainer

Dependencies

The meta-freescale-ml layer depends upon:

Recipe name Version Description
arm-compute-library 22.05 The ARM Computer Vision and Machine Learning library
armnn-swig 4.0.2 SWIG - Simplified Wrapper and Interface Generator
deepview-rt 2.4.46-aarch64 This package includes the updated and experimental ModelRunner for TensorFlow Lite and ARM NN. Also in this repository is a pre-release of DeepViewRT with support for the OpenVX backend.
deepview-rt-examples 1.6 DeepViewRT examples
nn-imx 1.3.0 i.MX Neural Networks Accelerator Plugin
nnstreamer 2.1.1 NNStreamer, Stream Pipeline Paradigm for Neural Network Applications
onnxruntime 1.10.0 cross-platform, high performance scoring engine for ML models
onnxruntime-native 1.10.0 Parts of ONNX Runtime that needs to be compiled for native system
packagegroup-fsl-ml 1.0 Add packages for AI/ML build
pytorch 1.9.1.post2 Tensors and Dynamic neural networks in Python with strong GPU acceleration
ssat 1.2.0+gitX Shell Script Automated Tester (unit testing executable files)
tensorflow-lite 2.9.1 TensorFlow Lite C++ Library
tensorflow-lite-host-tools 2.9.1 Host tools required for build of TensorFlow Lite C++ Library unit tests and Evaluation Tools
tensorflow-lite-vx-delegate 2.9.1 TensorFlow Lite VX Delegate
tensorflow-protobuf 3.9.2 Protocol Buffers - structured data serialisation mechanism
tim-vx 1.1.42 Tensor Interface Module for OpenVX
torchvision 0.10.0 The torchvision package consists of popular datasets, model architectures, and common image transformations for computer vision.
tvm 0.7.0 Open Deep Learning Compiler Stack