ML Workflow Automation for Training-to-Scoring Transition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The process of transforming a machine learning training experiment into a scoring experiment and subsequently into a web service is complex and not intuitive for users, requiring multiple distinct steps.
Innovation Solution
A system that automatically detects the need for an additional machine learning experiment, reconfigures the workflow, and consolidates elements to create a scoring experiment, reducing user interaction and streamlining the process by identifying necessary modules and connecting them properly, allowing for the creation of a web service with minimal user effort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the system provides detailed manual control over transforming training experiments into scoring experiments and web services, then users have full flexibility and control, but the process becomes complex and requires multiple distinct steps that are not intuitive
Solution Approach 1:
The system performs automatic detection and reconfiguration of experiment workflows. When a training experiment is successfully run, the system automatically detects that a scoring experiment may be needed and reconfigures the workflow by identifying and connecting necessary modules, reducing manual user intervention while maintaining transformation capability
Solution Approach 2:
The system automatically prompts users that an additional experiment workflow may be needed based on specific criteria associated with the first experiment workflow, before the user explicitly requests it. This preliminary detection and preparation simplifies the subsequent transformation process
2Productivity
If the system requires users to manually perform each step of transforming training experiments into scoring experiments and web services, then users have full control over the process, but user efficiency decreases due to increased mental effort and interaction steps
Solution Approach 1:
The system automatically reconfigures the first experiment workflow to perform the intended second experiment workflow by automatically identifying all necessary modules and connecting them properly, eliminating the need for users to manually perform each transformation step and significantly reducing interaction time
Solution Approach 2:
The system displays to the user the first experimental workflow transitioning from the first experiment workflow to the additional experiment workflow, providing visual feedback that confirms the automatic transformations and allows users to verify the process while minimizing their active participation
Data Source
AI summary
Automatically detecting and anticipating that an additional machine learning experiment may be needed. A method includes after successfully running a first experiment workflow, automatically prompting a user that an additional experiment workflow may be needed based on specific criteria associated with the first experiment workflow. The method further includes receiving input from the user confirming the additional experiment workflow. As a result of receiving input from the user confirming the additional experiment workflow, the method further includes the system automatically reconfiguring the first experiment workflow, including automatically identifying all necessary modules for the additional experiment workflow and connecting them properly to perform the intended second experiment workflow. The method further includes displaying to the user the first experimental workflow transitioning from the first experiment workflow to the additional experiment workflow.