Group By and Aggregation Functionality Available
Group By and Aggregation functions have been added to the Flow Studio to help you group your data in meaningful ways.
You can access all the aggregation functions in a convenient, all-in-one function interface within the Flow Studio.
Aggregation features available
|Group By||Creates groups from identical data values in a selected field.|
|Concat||Returns a string of the concatenated values in a group.|
|Count||Returns the count of the number of field values of a group.|
|First||Returns the first value in the value field from the group.|
|Last||Returns the last value in the value field from the group.|
|List||Returns a list of the concatenated values in a group.|
|Sum||Returns the sum value of a field in a group.|
Robin, your virtual assistant, keeps people in the loop by empowering them to make decisions for automated Flows. You can create tasks that Robin sends to your business users for them to review and approve.
A Better Robin Experience
Previously, users completed Robin's tasks one at a time in the order the Flow displayed them. Now, Robin gives users a list of all the tasks and lets them choose the tasks they think are most important.
This update expands Robin's capabilities for additional use cases, such as up-sell & cross-sell recommendations, in which your users need to get an overview of the tasks to assess their related priority before diving into the task details for review. To help with prioritizing tasks, table sorting and filtering capabilities are available.
Robin's new UX also provides a better degree of focus on the task list or task details, so your users are less distracted with too much information from other tasks.
Automation designers can select which end users (including users from your internal Enterprise authentication providers i.e. LDAP/AD) can have access to each Robin Skill. This ensures business users only see and work on the tasks they were assigned.
AI Studio Updates
Cancel a Model in Training
You can now quickly cancel a model training with a single click. If the model's performance you are working on doesn't look promising, dismiss that model version entirely.
Optimized Dense Layer Defaults
The defaults for the kernel and bias initializer configuration settings have been updated to be better optimized for intent detection models.
New Saving/Discarding/Recovering Process
We have updated the reminders to save, discard or recover AI Models to be the same as those in Flow Studio. You can now recover an AI model in the event you accidentally click away from AI Studio.
Extract Images from a PDF
Images are valuable data in PDF files. You can use the new function PDFExtractImage to extract images from PDF files in a Flow.