Merge branch 'optimize-extraction' into return-data-only
This commit is contained in:
commit
0496ddcb92
3 changed files with 6 additions and 9 deletions
|
|
@ -125,7 +125,7 @@ ENABLE_LLM_CACHE_FOR_EXTRACT=true
|
||||||
SUMMARY_LANGUAGE=English
|
SUMMARY_LANGUAGE=English
|
||||||
|
|
||||||
### Entity types that the LLM will attempt to recognize
|
### Entity types that the LLM will attempt to recognize
|
||||||
# ENTITY_TYPES='["Organization", "Person", "Location", "Event", "Technology", "Equipment", "Product", "Document", "Category"]'
|
# ENTITY_TYPES='["Person", "Organization", "Location", "Event", "Artifact", "CreativeWork"]'
|
||||||
|
|
||||||
### Chunk size for document splitting, 500~1500 is recommended
|
### Chunk size for document splitting, 500~1500 is recommended
|
||||||
# CHUNK_SIZE=1200
|
# CHUNK_SIZE=1200
|
||||||
|
|
|
||||||
|
|
@ -24,15 +24,12 @@ DEFAULT_SUMMARY_LENGTH_RECOMMENDED = 600
|
||||||
DEFAULT_SUMMARY_CONTEXT_SIZE = 12000
|
DEFAULT_SUMMARY_CONTEXT_SIZE = 12000
|
||||||
# Default entities to extract if ENTITY_TYPES is not specified in .env
|
# Default entities to extract if ENTITY_TYPES is not specified in .env
|
||||||
DEFAULT_ENTITY_TYPES = [
|
DEFAULT_ENTITY_TYPES = [
|
||||||
"Organization",
|
|
||||||
"Person",
|
"Person",
|
||||||
|
"Organization",
|
||||||
"Location",
|
"Location",
|
||||||
"Event",
|
"Event",
|
||||||
"Technology",
|
"Artifact",
|
||||||
"Equipment",
|
"CreativeWork",
|
||||||
"Product",
|
|
||||||
"Document",
|
|
||||||
"Category",
|
|
||||||
]
|
]
|
||||||
|
|
||||||
# Separator for graph fields
|
# Separator for graph fields
|
||||||
|
|
|
||||||
|
|
@ -323,7 +323,7 @@ async def _handle_single_entity_extraction(
|
||||||
if len(record_attributes) != 4 or "entity" not in record_attributes[0]:
|
if len(record_attributes) != 4 or "entity" not in record_attributes[0]:
|
||||||
if len(record_attributes) > 1 and "entity" in record_attributes[0]:
|
if len(record_attributes) > 1 and "entity" in record_attributes[0]:
|
||||||
logger.warning(
|
logger.warning(
|
||||||
f"{chunk_key}: extraction failed! only got {len(record_attributes)} feilds on entity `{record_attributes[1]}`"
|
f"{chunk_key}: extraction failed! Found {len(record_attributes)}/4 feilds on ENTITY `{record_attributes[1]}`"
|
||||||
)
|
)
|
||||||
return None
|
return None
|
||||||
|
|
||||||
|
|
@ -394,7 +394,7 @@ async def _handle_single_relationship_extraction(
|
||||||
if len(record_attributes) != 5 or "relationship" not in record_attributes[0]:
|
if len(record_attributes) != 5 or "relationship" not in record_attributes[0]:
|
||||||
if len(record_attributes) > 1 and "relationship" in record_attributes[0]:
|
if len(record_attributes) > 1 and "relationship" in record_attributes[0]:
|
||||||
logger.warning(
|
logger.warning(
|
||||||
f"{chunk_key}: extraction failed! only got {len(record_attributes)} fields on realtion `{record_attributes[1]}`"
|
f"{chunk_key}: extraction failed! Found {len(record_attributes)}/5 fields on REALTION `{record_attributes[1]}`"
|
||||||
)
|
)
|
||||||
return None
|
return None
|
||||||
|
|
||||||
|
|
|
||||||
Loading…
Add table
Reference in a new issue