ONNX
ONNX (Open Neural Network Exchange) is an open standard jointly initiated by Microsoft and Facebook that serves as a common intermediate format for representing machine learning models. It defines a unified operator set and data type specification, enabling models to be converted between different training frameworks such as PyTorch and TensorFlow, and deployed across diverse hardware platforms via inference engines such as ONNX Runtime, thereby addressing interoperability challenges in the deep learning ecosystem.
Core Facts
ONNX Runtime is an open-source machine learning inference and training acceleration engine primarily developed under Microsoft's leadership.
ONNX was originally co-initiated by Microsoft and Facebook (now Meta) in 2017.
Transformers.js is a JavaScript library developed by Hugging Face that, based on WebAssembly and WebGPU technology, ports Transformer model inference to browser environments.
Timeline (last 90 days)
ONNX (Open Neural Network Exchange) is an open model representation format designed so that models exported from different training frameworks (such as PyTorch and TensorFlow) can be loaded by a variety of inference engines.
An Indian automotive embedded software engineer with about one year of experience, whose tech stack covers C/C++, Python, embedded systems, ARM, RTOS, and Linux fundamentals, is learning machine learning and deep learning basics and plans to build hands-on projects around ASR/TTS, model deployment, ONNX, quantization, and edge hardware inference.
Vector normalization parameters still need to be saved in a separate file, because if the normalization layer is not also solidified into the computation graph, the inference side needs to save and load these statistics separately, otherwise observation distribution shift will cause the policy to fail.
After training is complete, the model is exported as an ONNX file for deployment; the authors added a TRAIN_CONFIG field to the ONNX file to record all training configurations for easy traceability.
BitLens is an open-source project that completes per-layer quantization sensitivity analysis with a single calibration and exports quantization configurations for mainstream frameworks.
BitLens wraps PyTorch nn.Module or ONNX computation graphs via hooks, collecting 128-bin histograms each for weights (symmetric per-channel) and activations (asymmetric per-tensor).
The open-source project onnx-wakeword provides inference code and examples for running ONNX wake word and keyword detection models in local applications.
ONNX is suited to desktop, Python, Android, and browser runtimes that support ONNX Runtime; TFLite or TFLite Micro is common on resource-constrained embedded devices.
When the network is unavailable, PaperToProfit falls back to the local ONNX model RapidOCR, detecting text boxes on the CPU and pairing dish names with quantities and prices via a two-dimensional spatial clustering algorithm that groups by a vertical midpoint threshold and sorts by horizontal coordinates.
Transformers.js底层使用ONNX Runtime Web，模型需要转为ONNX格式并量化，4-bit量化的1B模型大小约1.2GB而非5GB。
38 more timeline events
All Facts (20)
ONNX Runtime is an open-source machine learning inference and training acceleration engine primarily developed under Microsoft's leadership.
75%VerifiedONNX was originally co-initiated by Microsoft and Facebook (now Meta) in 2017.
70%VerifiedTransformers.js is a JavaScript library developed by Hugging Face that, based on WebAssembly and WebGPU technology, ports Transformer model inference to browser environments.
70%VerifiedONNX Runtime Web is the browser version of Microsoft's cross-platform inference engine, capable of directly loading standard ONNX format models
65%VerifiedThe ONNX (Open Neural Network Exchange) format allows PyTorch/TensorFlow models to be exported as a cross-platform intermediate representation format.
65%UnverifiedMoS-TTS-Nano使用ONNX（Open Neural Network Exchange）格式进行CPU上的高效推理
95%UnverifiedTensorFlow导出ONNX需要使用tf2onnx库
95%UnverifiedTransformers框架同时支持PyTorch和TensorFlow后端，并通过ONNX等格式支持更广泛的部署场景
90%UnverifiedONNX模型导出后，只要保持相同的input_names和output_names，推理代码就是完全通用的，无论模型来自PyTorch还是TensorFlow
90%UnverifiedTensorFlow导出ONNX时不需要提供dummy数据，因为tf.function通过AutoGraph将Python代码编译为静态图，input_signature提供的类型和形状信息足以完整描述计算图结构
90%Unverifiedonnxruntime-genai包可以用于ONNX格式模型的生成式AI推理
85%Unverified2019年，ONNX被移交给Linux基金会旗下的LF AI & Data基金会管理
85%UnverifiedONNX（Open Neural Network Exchange）诞生于2017年，最初由Facebook（Meta）和Microsoft联合开发
85%Unverified许多HuggingFace模型仓库已经提供了ONNX版本，特别是针对CPU和移动端优化的量化模型
80%UnverifiedONNX已获得Amazon、Intel、NVIDIA、Qualcomm等数十家主流硬件和软件厂商的支持
80%Partially VerifiedONNX由微软和Facebook于2017年联合发起，目前已成为AI模型互操作的事实标准
75%UnverifiedTensorRT是NVIDIA推出的高性能深度学习推理优化器与运行时，自2015年发布
70%UnverifiedAn Indian automotive embedded software engineer with about one year of experience, whose tech stack covers C/C++, Python, embedded systems, ARM, RTOS, and Linux fundamentals, is learning machine learning and deep learning basics and plans to build hands-on projects around ASR/TTS, model deployment, ONNX, quantization, and edge hardware inference.
50%UnverifiedONNX (Open Neural Network Exchange) is an open model representation format designed so that models exported from different training frameworks (such as PyTorch and TensorFlow) can be loaded by a variety of inference engines.
50%UnverifiedAfter training is complete, the model is exported as an ONNX file for deployment; the authors added a TRAIN_CONFIG field to the ONNX file to record all training configurations for easy traceability.
50%