In this post, I’ll show you how to optimize models in TensorFlow by using TensorRT for ONNX on Microsoft Azure.

Azure Machine Learning Hardware Accelerated Models (Project Brainwave) provides hardware accelerated machine learning with FPGA.
In Github tutorial, there are several useful helper classes and functions (with python) which encapsulate boilerplate code to achieve provisioning steps. In this post I show you the same steps without these helpers. With these steps I hope it helps you to understand new FPGA-enabled services and how it’s working.

Here I show you TensorFlowOnSpark on Azure Databricks. With this tutorial, you can learn how to use Azure Databricks through lifecycle, such as – cluster management, analytics by notebook, working with external libraries, working with surrounding Azure services, submitting a job for production, etc.

By using aztk, you can easily deploy and drop your Spark cluster in the cloud (Azure) and you can take agility for parallel programming (for ex, starting with low-capacity VMs, performance testing with large size or GPU accelerated, etc) with massive cloud computing power.
Here I show you our machine learning tutorials (PySpark and MLlib) with aztk.

In this post I show you Azure Batch AI fundamentals (how to use and how it works) using Azure CLI. You can find that Batch AI significantly simplifies your distributed training with Azure infrastructure.

The advantage of CNTK (Cognitive Toolkit) is not only performance, but it can also support multiple GPUs on multiple machines with built-in capabilities and rapidly scale along with the number of nodes and GPUs. Here I show you the step-by-step walkthrough for the distributed training with CNTK.

In this post I show you a walkthrough (tutorials and general tasks) of Graph database with Azure Cosmos DB Gremlin for your first use, and a little bit dive into the practical usage of the graph query.