grischperl
commented
1 month ago Evaluation Test results:
Run 1
************************************************************************
*** Test Results per Category:
************************************************************************
+------------------------+-----------+
| Category Name | Score (%) |
+------------------------+-----------+
| kubernetes | 16.33 |
| problem_finding | 52.63 |
| solution_finding | 4.1 |
| definition/explanation | 0.0 |
| step-by-step_guidance | 15.15 |
| yaml | 0.0 |
| completeness | 0.0 |
| kyma | 13.04 |
+------------------------+-----------+
************************************************************************
*** Overall success score across all expectations: 14.07%
************************************************************************Run 2
************************************************************************
*** Test Results per Category:
************************************************************************
+------------------------+-----------+
| Category Name | Score (%) |
+------------------------+-----------+
| kubernetes | 20.41 |
| problem_finding | 63.16 |
| solution_finding | 6.15 |
| definition/explanation | 0.0 |
| step-by-step_guidance | 27.27 |
| yaml | 0.0 |
| completeness | 0.0 |
| kyma | 13.04 |
+------------------------+-----------+
************************************************************************
*** Overall success score across all expectations: 17.78%
************************************************************************Run 3
************************************************************************
*** Test Results per Category:
************************************************************************
+------------------------+-----------+
| Category Name | Score (%) |
+------------------------+-----------+
| kubernetes | 20.41 |
| problem_finding | 52.63 |
| solution_finding | 7.69 |
| definition/explanation | 0.0 |
| step-by-step_guidance | 18.18 |
| yaml | 2.33 |
| completeness | 0.0 |
| kyma | 13.04 |
+------------------------+-----------+
************************************************************************
*** Overall success score across all expectations: 16.67%
************************************************************************
Description
Unite the Supervisor together with Finalizer and Planner under one subgraph. The subgraph is now called the Supervisor, with three nodes: Finalizer, Planner and Router. The newly introduced Router simply routes the task to the assigned agent.
Changes proposed in this pull request:
Also take a look at the following comment for a comprehensive To-Do list implemented in this PR: Comment #170
Related issue(s)
170