Openscale

Showing 28

Feature enhancement to monitor object detection models.

Openscale allows us to monitor the quality of machine learning models for classification, regression, and time series.There is a valuable opportunity to extend its capabilities to include monitoring and explainability for object detection models. ...

7 months ago in Openscale 0 Not under consideration

14 VOTE

WMLA Client Python Lib

Currently, WMLA has multiple ways to use and customers have a steep learning curve. WMLA REST API supports all the different CP4D versions and platforms. To reduce the learning curve and also reduce the workload to support different ways to use WM...

over 2 years ago in Openscale 0 Not under consideration

11 VOTE

WMLA EDT: Customization, Minimal Example & Documentation

Client asks how to use and customize the elastic distributed training (EDT) feature of WML-A. They explicitly complained about the lack of a detailed documentation (being aware of WMLA/Spectrum Conductor documentation in the knowledge center &...

over 3 years ago in Openscale 0 Future consideration

9 VOTE

Openscale Authoring HTTPS API Enhancement

Currently, we face the following weaknesses on OpenScale usage considering an MLOps pipeline automation use cases that require extensive use of OpenScale authoring API for automated monitoring of models within a deployment workflow:- Lack of prope...

5 months ago in Openscale 0 Not under consideration

8 VOTE

Implement multiple warnings prior to altering the database schema in the system setup section. Additionally, provide the capability to revert to the previous database schema, enabling users to retrieve prior evaluations.

Utility: The significance lies in the fact that modifying the schema currently results in the loss of all prior evaluations, without an option for retrieval. This enhancement aims to safeguard time and projects from potential loss, particularly in...

23 days ago in Openscale 0 Submitted

8 VOTE

Watchdog/Timeout option for jobs, so jobs that run longer than x minutes are stopped

There are situations where WML and /or spark jobs are hanging/ running to long. My CP4D job developer have a good understanding of how long a job should run. In case of longer running jobs I as a CP4D administrator want that these jobs are aborted...

over 1 year ago in Openscale 0 Not under consideration

3 VOTE

Grafana / Prometheus GPU metric in CP4D

It would be nice to have the GPU utilization metrics used in WMLA also introduced to the main monitoring dashboard of core CP4D. This is so that we still have a single monitoring dashboard for all services of CP4D.

over 2 years ago in Openscale 0 Future consideration

3 VOTE

Audit logs for Watson Assistant.

My dev team can't keep track of who and which changes where made on the chatbot.It would be nice having a change history with version date, change sets and the user who made those changes.

over 5 years ago in Openscale 2 Not under consideration

2 VOTE

To include unsupervised (clustering) models to Watson Openscale/ Trust

9 months ago in Openscale 0 Future consideration

2 VOTE

Tensorboard support on job monitoring

TensorBoard is TensorFlow’s visualization toolkit enables tracking metrics like loss and accuracy, visualize the model graph, view histograms of weights, biases, or other tensors as they change over time, and much more. A managed TensorBoard suppo...

over 2 years ago in Openscale 0 Future consideration

Please enter your email address

FILTER BY CATEGORY