Type: nodetool.document.SplitSentences
Namespace: nodetool.document
Description
Splits text into sentences using LangChain’s SentenceTransformersTokenTextSplitter. sentences, split, nlp
Use cases:
- Natural sentence-based text splitting
- Creating semantically meaningful chunks
- Processing text for sentence-level analysis
Properties
| Property | Type | Description | Default |
|---|---|---|---|
| document | document |
{'type': 'document', 'uri': '', 'asset_id': None, 'data': None} |
|
| chunk_size | int |
Maximum number of tokens per chunk | 40 |
| chunk_overlap | int |
Number of tokens to overlap between chunks | 5 |
Outputs
| Output | Type | Description |
|---|---|---|
| text | str |
|
| source_id | str |
|
| start_index | int |
Metadata
Related Nodes
Browse other nodes in the nodetool.document namespace.