zhuyongxin
7ae9707a3b
Persist provider reasoning and assistant text separately on agent_reasoning_audit (DeepSeekAssistantMessage path), extract diagnosis_run.conclusion, enrich RAG tool audit (step_id/query/qualityScore), gate empty mysql tools, drop devtools, and align MVP docs after live E2E verification.
43 lines
1.2 KiB
PowerShell
43 lines
1.2 KiB
PowerShell
param(
|
|
[string]$Cases = "eval\rag-retrieval\cases\golden-cases.json",
|
|
[string]$Fixtures = "eval\rag-retrieval\fixtures",
|
|
[string]$RetrievedAt = "",
|
|
[string]$KbScope = "rag-eval",
|
|
# hybrid = production main path; dense = same-collection recall baseline
|
|
[ValidateSet("hybrid", "dense")]
|
|
[string]$SearchMode = "hybrid",
|
|
[switch]$SkipEval
|
|
)
|
|
|
|
$ErrorActionPreference = "Stop"
|
|
|
|
$mavenArgs = @(
|
|
"-q",
|
|
"-Dtest=RagLookupSnapshotGeneratorTest",
|
|
"-Drag.snapshot.enabled=true",
|
|
"-Drag.snapshot.cases=$Cases",
|
|
"-Drag.snapshot.fixtures=$Fixtures",
|
|
"-Dretrieval.kb-scope=$KbScope",
|
|
"-Dretrieval.search.mode=$SearchMode"
|
|
)
|
|
|
|
if ($RetrievedAt -ne "") {
|
|
$mavenArgs += "-Drag.snapshot.retrievedAt=$RetrievedAt"
|
|
}
|
|
|
|
$mavenArgs += "test"
|
|
|
|
Write-Host "Generating RAG lookupResult fixtures (search.mode=$SearchMode, kb-scope=$KbScope)..."
|
|
& mvn @mavenArgs
|
|
if ($LASTEXITCODE -ne 0) {
|
|
throw "Fixture generation failed with exit code $LASTEXITCODE"
|
|
}
|
|
|
|
if (-not $SkipEval) {
|
|
Write-Host "Running offline RAG retrieval baseline..."
|
|
& python scripts\eval_rag_retrieval.py --cases $Cases --fixtures $Fixtures
|
|
if ($LASTEXITCODE -ne 0) {
|
|
throw "Offline eval failed with exit code $LASTEXITCODE"
|
|
}
|
|
}
|