Update README.md
Browse files
README.md
CHANGED
|
@@ -7,6 +7,8 @@ pipeline_tag: token-classification
|
|
| 7 |
---
|
| 8 |
# bert-chunker-3
|
| 9 |
|
|
|
|
|
|
|
| 10 |
bert-chunker-3 is a text chunker based on BertForTokenClassification to predict the start token of chunks (for use in RAG, etc), and using a sliding window it cuts documents of any size into chunks. We see it as an alternative of [semantic chunker](https://github.com/FullStackRetrieval-com/RetrievalTutorials/blob/main/tutorials/LevelsOfTextSplitting/5_Levels_Of_Text_Splitting.ipynb), but specially, it not only works for the structured texts, but also the **unstructured and messy texts**.
|
| 11 |
|
| 12 |
Different from [bc-2](https://huggingface.co/tim1900/bert-chunker-2) and [bc](https://huggingface.co/tim1900/bert-chunker), to overcome the data distribution shift, our training data were labeled by a LLM and trainng pipeline was improved, therefore it is **more stable**.
|
|
|
|
| 7 |
---
|
| 8 |
# bert-chunker-3
|
| 9 |
|
| 10 |
+
[GitHub](https://github.com/jackfsuia/bert-chunker/tree/main/bc3)
|
| 11 |
+
|
| 12 |
bert-chunker-3 is a text chunker based on BertForTokenClassification to predict the start token of chunks (for use in RAG, etc), and using a sliding window it cuts documents of any size into chunks. We see it as an alternative of [semantic chunker](https://github.com/FullStackRetrieval-com/RetrievalTutorials/blob/main/tutorials/LevelsOfTextSplitting/5_Levels_Of_Text_Splitting.ipynb), but specially, it not only works for the structured texts, but also the **unstructured and messy texts**.
|
| 13 |
|
| 14 |
Different from [bc-2](https://huggingface.co/tim1900/bert-chunker-2) and [bc](https://huggingface.co/tim1900/bert-chunker), to overcome the data distribution shift, our training data were labeled by a LLM and trainng pipeline was improved, therefore it is **more stable**.
|