_posts/2018-05-24-mxnet1304.markdown - infrastructure-blogs-archive - Git at Google

 ---
 layout: post
 title: Apache MXNet 1.2.0 Release is out!
 date: '2018-05-24T23:03:26+00:00'
 categories: mxnet
 ---
 <p>
 Today Apache MXNet community announced the 1.2 release of the Apache MXNet deep learning framework. The new capabilities in MXNet provide the following benefits to users:
 <ol>
 <li><i>MXNet is easier to use</i></li>
 <ul><li><b>New scala inference APIs</b>: This release includes new Scala inference APIs which offer an easy-to-use, Scala idiomatic and thread-safe high level APIs for performing predictions with deep learning models trained with MXNet.</li>
 <li><b>Exception Handling Support for Operators</b>: MXNet now transports backend C++ exceptions to the different language front-ends and prevents crashes when exceptions are thrown during operator execution</li></ul>
 <li><i>MXNet is faster</i></li>
 <ul><li><b>MKL-DNN integration</b>: MXNet now integrates with Intel MKL-DNN to accelerate neural network operators: Convolution, Deconvolution, FullyConnected, Pooling, Batch Normalization, Activation, LRN, Softmax, as well as some common operators: sum and concat. This integration allows NDArray to contain data with MKL-DNN layouts and reduces data layout conversion to get the maximal performance from MKL-DNN. Currently, the MKL-DNN integration is still experimental. </li>
 <li><b>Enhanced FP16 support</b>: MXNet now adds support for distributed mixed precision training with FP16. It supports storing of master copy of weights in float32 with the multi_precision mode of optimizers. Improved speed of float16 operations on x86 CPU by 8 times through F16C instruction set. </li></ul>
 <li><i>MXNet provides easy interoperability</i></li>
 <ul><li><b>Import ONNX models into MXNet</b>: Implemented a new ONNX module in MXNet which offers an easy to use API to import ONNX models into MXNet's symbolic interface. Checkout the <a href="https://github.com/apache/incubator-mxnet/blob/master/example/onnx/super_resolution.py">example</a> on how you could use this <a href="https://cwiki.apache.org/confluence/display/MXNET/ONNX-MXNet+API+Design">API</a> to import ONNX models and perform inference on MXNet. Currently, the ONNX-MXNet Import module is still experimental.</li></ul>
 </ol>

 <h3>Getting started with MXNet</h3>
 Getting started with <a href="http://mxnet.incubator.apache.org/install/index.html">MXNet</a> is simple. To learn more about the Gluon interface and deep learning, you can reference this <a href="http://gluon.mxnet.io/">comprehensive set of tutorials</a>, which covers everything from an introduction to deep learning to how to implement cutting-edge neural network models. If you’re a contributor to a machine learning framework, check out the interface specs on <a href="https://github.com/gluon-api/gluon-api/">GitHub</a>.
	---
	layout: post
	title: Apache MXNet 1.2.0 Release is out!
	date: '2018-05-24T23:03:26+00:00'
	categories: mxnet
	---
	<p>
	Today Apache MXNet community announced the 1.2 release of the Apache MXNet deep learning framework. The new capabilities in MXNet provide the following benefits to users:
	<ol>
	<li><i>MXNet is easier to use</i></li>
	<ul><li><b>New scala inference APIs</b>: This release includes new Scala inference APIs which offer an easy-to-use, Scala idiomatic and thread-safe high level APIs for performing predictions with deep learning models trained with MXNet.</li>
	<li><b>Exception Handling Support for Operators</b>: MXNet now transports backend C++ exceptions to the different language front-ends and prevents crashes when exceptions are thrown during operator execution</li></ul>
	<li><i>MXNet is faster</i></li>
	<ul><li><b>MKL-DNN integration</b>: MXNet now integrates with Intel MKL-DNN to accelerate neural network operators: Convolution, Deconvolution, FullyConnected, Pooling, Batch Normalization, Activation, LRN, Softmax, as well as some common operators: sum and concat. This integration allows NDArray to contain data with MKL-DNN layouts and reduces data layout conversion to get the maximal performance from MKL-DNN. Currently, the MKL-DNN integration is still experimental. </li>
	<li><b>Enhanced FP16 support</b>: MXNet now adds support for distributed mixed precision training with FP16. It supports storing of master copy of weights in float32 with the multi_precision mode of optimizers. Improved speed of float16 operations on x86 CPU by 8 times through F16C instruction set. </li></ul>
	<li><i>MXNet provides easy interoperability</i></li>
	<ul><li><b>Import ONNX models into MXNet</b>: Implemented a new ONNX module in MXNet which offers an easy to use API to import ONNX models into MXNet's symbolic interface. Checkout the <a href="https://github.com/apache/incubator-mxnet/blob/master/example/onnx/super_resolution.py">example</a> on how you could use this <a href="https://cwiki.apache.org/confluence/display/MXNET/ONNX-MXNet+API+Design">API</a> to import ONNX models and perform inference on MXNet. Currently, the ONNX-MXNet Import module is still experimental.</li></ul>
	</ol>

	<h3>Getting started with MXNet</h3>
	Getting started with <a href="http://mxnet.incubator.apache.org/install/index.html">MXNet</a> is simple. To learn more about the Gluon interface and deep learning, you can reference this <a href="http://gluon.mxnet.io/">comprehensive set of tutorials</a>, which covers everything from an introduction to deep learning to how to implement cutting-edge neural network models. If you’re a contributor to a machine learning framework, check out the interface specs on <a href="https://github.com/gluon-api/gluon-api/">GitHub</a>.