Machine learning is a vast area and covers several mechanisms of learning such as supervised learning, unsupervised learning, deep learning etc. As a result, there are several frameworks available which include TensorFlow from Google Brain team and Apache MxNet.
What is MxNet?
Apache MxNet is a deep learning framework that is open source and is used to train and deploy deep neural networks. Deep Learning refers to algorithms in Machine learning where data is passed through the training model multiple times. Each time, the output from the previous pass is used as input for next pass. Apache MxNet allows researchers to train the models using deep learning. Moreover, it can also scale to multiple GPUs and multiple machines.
Characteristic Features
- Scalable – As mentioned above, the framework can use multiple GPUs across multiple servers. It can be easily deployed in the cloud to take advantage of the elasticity of the resources.
- Flexible – Developers can use imperative (procedural) programming language such as C/C++ or symbolic programming languages such as LISP or ProLog.
- Multiple language support – C/C++ for the backend to get most out of the GPUs and Python, Scala etc for the frontend for developers.
- Portable – the models trained using MxNet could be deployed on a variety of devices such as mobiles, IoT devices or containers. They could also be utilized in serverless computing. These devices typically have weaker CPU and/or limited memory.
Cloud support:
This framework is readily supported by major cloud providers such as AWS and Azure.
Alternatives
One of the closest competitors is TensorFlow, which was developed and released by Google Brain team. Followers of these frameworks feel that TensorFlow is more research-oriented whereas MxNet is for production purposes.
Related Links
Related Keywords
Machine Learning, Deep Learning, Supervised Learning, Unsupervised Learning, TensorFlow, Gluon