ModelAnalysis represents the analysis results for a single model
| 29 | |
| 30 | // ModelAnalysis represents the analysis results for a single model |
| 31 | type ModelAnalysis struct { |
| 32 | ModelName string `json:"model_name"` |
| 33 | BatchSource string `json:"batch_source"` // Which batch directory this model came from |
| 34 | ToolInvocation MetricSet `json:"tool_invocation"` // Binary: should call tool vs did call tool |
| 35 | ToolSelection MetricSet `json:"tool_selection"` // Specific: right tool vs wrong tool |
| 36 | AverageResponseTime float64 `json:"average_response_time"` // Average response time in seconds |
| 37 | TotalTests int `json:"total_tests"` |
| 38 | TotalRuns int `json:"total_runs"` |
| 39 | ResultFiles []string `json:"result_files"` |
| 40 | } |
| 41 | |
| 42 | // BatchAnalysisReport represents the complete analysis report |
| 43 | type BatchAnalysisReport struct { |
nothing calls this directly
no outgoing calls
no test coverage detected