mirror of
https://github.com/trustgraph-ai/trustgraph.git
synced 2026-05-11 00:02:37 +02:00
Fix incorrect property names in explainability (#698)
Remove type suffixes from explainability dataclass fields + fix show_explain_trace Rename dataclass fields to match KG property naming conventions: - Analysis: thought_uri/observation_uri → thought/observation - Synthesis/Conclusion/Reflection: document_uri → document Fix show_explain_trace for current API: - Resolve document content via librarian fetch instead of removed inline content fields (synthesis.content, conclusion.answer) - Add Grounding display for DocRAG traces - Update fetch_docrag_trace chain: Question → Grounding → Exploration → Synthesis - Pass api/explain_client to all print functions for content resolution Update all CLI tools and tests for renamed fields.
This commit is contained in:
parent
a115ec06ab
commit
c387670944
6 changed files with 122 additions and 62 deletions
|
|
@ -728,8 +728,8 @@ def _question_explainable_api(
|
|||
|
||||
elif isinstance(entity, Synthesis):
|
||||
print(f"\n [synthesis] {prov_id}", file=sys.stderr)
|
||||
if entity.document_uri:
|
||||
print(f" Document: {entity.document_uri}", file=sys.stderr)
|
||||
if entity.document:
|
||||
print(f" Document: {entity.document}", file=sys.stderr)
|
||||
|
||||
else:
|
||||
if debug:
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue