About ML Skills

An ML skill is a consumer-ready, live deployment of an ML or OS package. Once deployed as ML skills, both ML Packages and OS Packages become models ready to be consumed within RPA workflows, simply by dragging and dropping an ML Skill Activity in UiPath Studio.

It is recommended that users acting as Process Controllers handle model deployments. More details about recommended user personas here.

The ML Skills page, accessible from the ML Skills menu after selecting a project, displays all the models deployed on your AI Center service, whether they use ML or OS packages. The page also displays their status, package name, version, whether or not a GPU is required, and prediction. Moreover, you can deploy packages by creating ML skills, delete them, view their detailed information, parameters, versions, or manage package versions - if available.

Basic monitoring is provided on skills in the form of prediction counts (number of requests to this skill).

The ML Skill Details page, accessible by selecting a skill from the list, contains detailed information about a skill, along with some version management actions.

Find information about how to manage your skills here.

Consuming ML Skills from Studio

To use a skill in a UiPath Studio Activity, the following environment must be set up:

• UiPath Studio v2019.10+, with UiPath Robot v2019.10+
• Installed UiPath.MLServices.Activities package from the Manage Packages menu in UiPath Studio.

Open UiPath Studio, drag and drop the ML Skill Activity into the RPA workflow and select the Refresh ML Skills option. This action will populate the drop-down list with all the successfully deployed skills from the Orchestrator connected to this Robot.

Pass the data to the input of the skill exactly as you would with any other activity. The skill also allows you to live-test an input by selecting Test Skill.

More documentation on the ML Skill Activity can be found here.

Watch the following video to learn how to use the Test Skill button to see what the deployed model is doing:

How are inputs passed to the activity

Depending on the input type, the MLSkills activity expects the following format:

JSON

"this is an example of input"
"{""expected-field"":""this is another example""}"

File

"C:/full/path/to/file.ext"

Files

"C:/full/path/to/file1.ext,C:/full/path/to/file2.ext,C:/full/path/to/file3.ext"