Industry News

Tensorflow  Open source machine learning platform

https://tensorflow.google.cn/

TensorFlow is an open source, Python-based machine learning framework. It was developed by Google and has rich applications in scenarios such as graph classification, audio processing, recommendation systems, and natural language processing. It is currently the most popular machine learning framework.

In addition to Python, TensorFlow also provides interfaces for other programming languages ​​such as C/C++, Java, Go, and R.

The open source deep learning library TensorFlow allows the deployment of deep neural network calculations to any number of CPUs or GPUs on servers, PCs, or mobile devices, leveraging only a single TensorFlow API. You may ask, there are many other deep learning libraries, such as Torch, Theano, Caffe and MxNet, so what is the difference between TensorFlow and other deep learning libraries? Most deep learning libraries, including TensorFlow, are capable of automatic derivation, are open source, support multiple CPU/GPUs, have pre-trained models, and support commonly used NN architectures, such as Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN), and Deep Belief Networks (DBN).

TensorFlow has more features, as follows:

• Supports all popular languages ​​like Python,C++、Java, R and Go.

• Can work on multiple platforms, even mobile and distributed platforms.

• It is supported by all cloud services (AWS, Google and Azure).

• Keras - a high-level neural network API, integrated with TensorFlow.

• Compared with Torch/Theano, TensorFlow has better calculation graph visualization.

• Allows the model to be deployed into industrial production and is easy to use.

• There is very good community support.

• TensorFlow is more than just a software library, it is a suite of software including TensorFlow, TensorBoard and TensorServing.