support images and tables extract from docx (#4619)
This commit is contained in:
parent
5893ebec55
commit
233c4150d1
10 changed files with 163 additions and 23 deletions
|
|
@ -428,7 +428,7 @@ class IndexingRunner:
|
|||
chunk_size=segmentation["max_tokens"],
|
||||
chunk_overlap=chunk_overlap,
|
||||
fixed_separator=separator,
|
||||
separators=["\n\n", "。", ".", " ", ""],
|
||||
separators=["\n\n", "。", ". ", " ", ""],
|
||||
embedding_model_instance=embedding_model_instance
|
||||
)
|
||||
else:
|
||||
|
|
@ -436,7 +436,7 @@ class IndexingRunner:
|
|||
character_splitter = EnhanceRecursiveCharacterTextSplitter.from_encoder(
|
||||
chunk_size=DatasetProcessRule.AUTOMATIC_RULES['segmentation']['max_tokens'],
|
||||
chunk_overlap=DatasetProcessRule.AUTOMATIC_RULES['segmentation']['chunk_overlap'],
|
||||
separators=["\n\n", "。", ".", " ", ""],
|
||||
separators=["\n\n", "。", ". ", " ", ""],
|
||||
embedding_model_instance=embedding_model_instance
|
||||
)
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue