IBM Watson Studio

IBM Watson Studio is the tools for scientists-analysts, application developers and profile specialists allowing to work jointly with data in the course of creation and training of models. The flexibility of this product allows to create models in which data can be placed and unrolled in any place in the hybrid environment that helps to master and apply a data science quicker.

2019

Start of sales on the MerliONCloud platform

On April 18, 2019 the company MERLION, Russian the VAD distributor, within development of the platform Merlioncloud, announces start of sales of two products - IBM SPSS Statistics Subscription and IBM Watson Studio Desktop Subscription. In more detail here.

Features

According to information for 2019 IBM Watson Studio provides:

• Data preparation
  • An opportunity to find the communications and patterns hidden in data using the built-in functions of cleaning and data translation of Data Refinery, to browse data in a tabular style, including visualization and summary statistics. It will help to state correctly requests to data and to receive exact and useful answers.

• Research of data
  • Using the built-in summary panels it is possible to create the data visualizations allowing to see earlier unnoticed patterns, interrelations and other aspects in real time. These visualization can share with the colleagues.

• Development of models
  • Using the configured computing environments scaled together with workflows it is convenient to test and unroll models. It is possible to select the optimal environment depending on a stage of modeling and scale. At choice Wednesdays Anaconda, Spark and GPU of different capacity are offered.

• Model assessment
  • Visualization of compliance of model and data with the help of the Model Visualization component entering into SPSS Modeler helps to make model more effective and more productive.

• Deployment of models
  • When the model is ready, it can be unrolled and estimated using service IBM Watson Machine Learning.

• Management of models
  • The Deep Learning Experiments component allows to compare runs of models and it is easy to optimize models at the level of hyper parameters.