Table of contents

What's new in Watson Studio Local

Check out what's new for Watson Studio Local!

What's new in Watson Studio Local Version 1.2.3.1 x86 - patch06 (August 2019)

Locate the following enhancements and fixes in Fix Central. Select the four wsl-x86-v1231-patch06-part0* files.

The following enhancements and fixes are provided in Patch 06.

Installation, security and IT administration
  • The Github/Bitbucket integration was enhanced to allow data scientists to create projects from a specific branch in a repository.
  • An admin can now control the creation of the Spark Context by default in Jupyter 2.7 and 3.5 environments.
  • The pyarrow and the compatible pandas package are included in Jupyter 2.7 and 3.5 environments.
  • The version of the sparkmagics package included with the Jupyter 2.7 and 3.5 environments is upgraded.
  • TS001906969 - All collaborators in a project are allowed to view the jobs that are scheduled by the project admins or editors.
  • Additional defect fixes. Learn more

What's new in Watson Studio Local Version 1.2.3.1 x86 - patch05 (July 2019)

Locate the following enhancements and fixes in Fix Central. Select wsl-x86-v1231-patch05.

The following enhancements and fixes are provided in patch05.

Installation, security, IT administration, and usage tracking
  • Automate some administration tasks such as user onboarding or asset management using the command line interface (CLI.) Learn more
  • Watson Studio Local administrators now have the ability to disable the sign up feature for users to request access to the Watson Studio cluster. Learn more
  • Additional defect fixes. Learn more

What's new in Watson Studio Local Version 1.2.3.1 x86 - patch04 (May 2019)

The backup and restore process is now enhanced. Go to Fix Central and download wsl-backup-restore.tar.gz. Learn more about backing up and restoring Watson Studio Local.

What's new in Watson Studio Local Version 1.2.3.1 x86 - patch02, patch03 (May 2019)

The following enhancements are available for notebooks and environments related to only GPU. Go to Fix Central and select both wsl-x86-patch02-TS002167674-TS002051550 and wsl-x86-v1231-patch03-TS002286373 to get the enhancements.

Notebooks and Environments (GPU only)

  • Kernel startup failure is now addressed.
  • Upgrades Python to version 3.6.
  • Anaconda is now included and supports both conda and pip-based package installations.
  • Upgrades the Nvidia CUDA toolkit to version 10.0 and the cuDNN library to 7.5.0.
  • TensorBoard is now integrated in the GPU image and allows data scientists to visualize TensorFlow graphs, images, and histograms.
  • Upgrades GPU-enabled frameworks that now include: TensorFlow, PyTorch, Keras, Theano, MXNet, LightGBM, and XGBoost.

What's new in Watson Studio Local Version 1.2.3.1 x86 - patch01 (April 2019)

Locate the following enhancements for data sources and managing images in Fix Central. Select wsl-x86-v1231-patch01-TS002078840-TS002078850 to get the enhancements.

Installation, security, IT administration, and usage tracking

  • Watson Studio Local administrators now have the option on whether or not to allow data scientists to create data sources with shared credentials. Learn more
  • Watson Studio Local administrators can now control the visibility of custom images created by data scientists. The administrators can specify only approved images that are visible to other data scientists. Learn more

What's new in Watson Studio Local Version 1.2.3.1 Power (April 2019)

Installation, security, IT administration, and usage tracking

  • The Watson Studio Local installation now supports silent installation.
  • The Watson Studio Local installation now supports new configuration parameters to secure the platform that allow users to optionally remove the sshd and netcat services from applicable pods.
  • The Watson Studio Local platform is further secured with updates to various components across all docker images to be free from known security vulnerabilities or exposures.

Explore, train, and deploy at scale using Hadoop Integration Service

Watson Studio Local now supports Cloudera Distribution for Hadoop version 5.16 and 6.0.1.

What's new in Watson Studio Local Version 1.2.3.1 x86 (March 2019)

Installation, security, IT administration, and usage tracking

  • The Watson Studio Local installation now supports new configuration parameters to secure the platform that allow users to optionally remove the sshd and netcat services from applicable pods.
  • The Watson Studio Local platform is further secured with updates to various components across all docker images to be free from known security vulnerabilities or exposures.

Explore, train, and deploy at scale using Hadoop Integration Service

  • Watson Studio Local now supports Cloudera Distribution for Hadoop version 5.16 and 6.0.1.

What's new in Watson Studio Local Version 1.2.3 (December 2018)

Installation, security, IT administration, and usage tracking

  • The Watson Studio Local wdp.conf installation now supports using your own NFS version 4.x server for storage. Learn more
  • The docker_redhat_install.sh script now supports overlay2 or devicemapper as the storage driver. Learn more

Security

  • Using SAML, an administrator can now set up single sign-on (SSO) to redirect Watson Studio Local users to sign in securely through your identity provider's login page. Learn more

Administration

  • The Admin Console now separates the alert settings and percentages for CPU usage, memory usage, and disk space usage.
  • In the Admin Console, an administrator can now search the pods list.
  • From the Scripts panel, an administrator has a new option in Add Spark jars in Watson Studio (moveJarClasspath.sh) to upload JAR files to the /user-home/_global_/spark/jars/ directory for use with Spark.

Usage tracking

  • An administrator can now retrieve daily usage reports as zipped CSV files by running the /wdp/utils/get_usage_reports.sh script. Learn more

Explore, train, and deploy at scale using Hadoop Integration Service

  • Users can now batch score and evaluate models on a Hadoop cluster, and leverage all model management capabilities in Watson Studio Local for the Hadoop models. Learn more
  • You can now call new utility methods in Watson Studio Local to save models on Hadoop and load them in Watson Studio Local. See the new sample Jupyter notebooks that show the usage. Learn more
  • Watson Studio Local provides new sample Jupyter notebooks that demonstrate how to work with HDP 3.0.1 Hive tables in the Spark catalog, and Hive catalog using the HiveWareHouseConnector with LLAP enabled.
  • Hadoop Integration gateway now works seamlessly with HA enabled WebHDFS.
  • A service user can now install the Hadoop registration service. Previously, only a root user could install it. Learn more
  • From the project Jobs tab, users can now run .R scripts natively on the Hadoop cluster Learn more

Data set changes

  • Watson Studio Local now enforces case sensitivity for the remote data sets Insert to code feature. Ensure that all schema and table names in your preexisting remote data sets match the exact case of the schema and table names in the data source, and correct any case mismatches in your remote data source definitions. It is also recommended that you browse to tables when creating remote data sets, to avoid accidentally typing in the wrong table name case. Learn more