onnxruntime-inference-examples

Inference framework

Repository providing examples for using ONNX Runtime (ORT) to perform machine learning inferencing.

Examples for using ONNX Runtime for machine learning inferencing.

GitHub

1k stars
38 watching
342 forks
Language: C++
last commit: almost 2 years ago

Related projects:

RepositoryDescriptionStars
microsoft/onnxruntime-training-examplesAccelerates training of large transformer models by providing optimized kernels and memory optimizations.317
microsoft/onnxruntimeA cross-platform, high-performance machine learning accelerator14,990
mlcommons/inferenceMeasures the performance of deep learning models in various deployment scenarios.1,256
xboot/libonnxAn onnx inference engine for embedded devices with hardware acceleration support589
alrevuelta/connxrAn embedded device-friendly C ONNX runtime with zero dependencies196
triton-inference-server/clientClient libraries and examples for communicating with Triton using various programming languages579
tpoisonooo/llama.onnxA project providing onnx models and tools for inference with LLaMa transformer model on various devices356
kraiskil/onnx2cGenerates C code from ONNX files for efficient neural network inference on microcontrollers234
emergentorder/onnx-scalaAn API and backend for running ONNX models in Scala 3 using typeful, functional deep learning and classical machine learning.138
sassoftware/enlighten-applyProvides code and materials to apply SAS machine learning techniques in software development127
nilseuropa/ros_ncnnA ROS wrapper for NCNN, allowing developers to integrate high-performance neural network inference into ROS projects63
oramasearch/onnx-goA Go package that allows developers to import pre-trained neural network models without being tied to a framework or library.726
roboflow/inferenceA platform for deploying and fine-tuning computer vision models in production-ready environments.1,401
microsoft/0xdeca10bA framework for hosting and training machine learning models on a blockchain, enabling secure sharing and prediction without requiring users to pay for data or model updates.559