onnx 深度学习
介绍(Introduction)
ONNX (Open Neural Network Exchange Format) is a format designed to represent any type of Machine Learning and Deep Learning model.
ONNX(开放神经网络交换格式)是一种旨在表示任何类型的机器学习和深度学习模型的格式。
Some example of supported frameworks are: PyTorch, TensorFlow, Keras, SAS, Matlab, and many more. In this way, ONNX can make it easier to convert models from one framework to another. Additionally, using ONNX.js we can then easily deploy online any model which has been saved in an ONNX format.
受支持框架的示例包括:PyTorch,TensorFlow,Keras,SAS,Matlab等。 这样,ONNX可以更轻松地将模型从一个框架转换为另一个框架。 此外,使用ONNX.js,我们可以轻松地在线部署已以ONNX格式保存的任何模型。
In Figure 1, is available a simple example of a Variational Autoencoder PyTorch model deployed online using ONNX.js in order to make inference on demand. A complete working example of this deployed model is available at this link on my personal website.
在图1中,提供了一个简单的示例,该示例使用ONNX.js在线部署了变分式自动编码器PyTorch模型,以便根据需要进行推理。 我个人网站上的此链接提供了此部署模型的完整工作示例。