-
Notifications
You must be signed in to change notification settings - Fork 22
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
3 changed files
with
133 additions
and
7 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,2 +1,3 @@ | ||
| from ._chroma import ChromaSourceStorage | ||
| from ._demo import RagnaDemoSourceStorage | ||
| from ._lancedb import LanceDBSourceStorage |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,117 @@ | ||
| from ragna.core import ( | ||
| Document, | ||
| PackageRequirement, | ||
| RagnaId, | ||
| Requirement, | ||
| Source, | ||
| SourceStorage, | ||
| ) | ||
|
|
||
| from ragna.utils import chunk_pages, page_numbers_to_str, take_sources_up_to_max_tokens | ||
|
|
||
|
|
||
| class LanceDBSourceStorage(SourceStorage): | ||
| @classmethod | ||
| def display_name(cls) -> str: | ||
| return "LanceDB" | ||
|
|
||
| @classmethod | ||
| def requirements(cls) -> list[Requirement]: | ||
| return [ | ||
| PackageRequirement("lancedb>=0.2"), | ||
| # FIXME: re-add this after https://github.com/apache/arrow/issues/38167 is | ||
| # resolved. | ||
| # PackageRequirement("pyarrow"), | ||
| PackageRequirement("sentence_transformers"), | ||
| ] | ||
|
|
||
| def __init__(self, config): | ||
| super().__init__(config) | ||
|
|
||
| import lancedb | ||
| import pyarrow as pa | ||
| from sentence_transformers import SentenceTransformer | ||
|
|
||
| self._db = lancedb.connect(config.local_cache_root / "lancedb") | ||
| self._model = SentenceTransformer("paraphrase-albert-small-v2") | ||
| self._schema = pa.schema( | ||
| [ | ||
| pa.field("document_id", pa.string()), | ||
| pa.field("document_name", pa.string()), | ||
| pa.field("page_numbers", pa.string()), | ||
| pa.field("text", pa.string()), | ||
| pa.field( | ||
| self._VECTOR_COLUMN_NAME, | ||
| pa.list_(pa.float32(), self._model[-1].word_embedding_dimension), | ||
| ), | ||
| pa.field("num_tokens", pa.int32()), | ||
| ] | ||
| ) | ||
|
|
||
| def _embed(self, batch): | ||
| return [self._model.encode(sentence) for sentence in batch] | ||
|
|
||
| _VECTOR_COLUMN_NAME = "embedded_text" | ||
|
|
||
| def store( | ||
| self, | ||
| documents: list[Document], | ||
| *, | ||
| chat_id: RagnaId, | ||
| chunk_size: int = 500, | ||
| chunk_overlap: int = 250, | ||
| ) -> None: | ||
| table = self._db.create_table(name=str(chat_id), schema=self._schema) | ||
|
|
||
| for document in documents: | ||
| for chunk in chunk_pages( | ||
| document.extract_pages(), | ||
| chunk_size=chunk_size, | ||
| chunk_overlap=chunk_overlap, | ||
| tokenizer=self._model.tokenizer, | ||
| ): | ||
| table.add( | ||
| [ | ||
| { | ||
| "document_id": str(document.id), | ||
| "document_name": document.name, | ||
| "page_numbers": page_numbers_to_str(chunk.page_numbers), | ||
| "text": chunk.text, | ||
| self._VECTOR_COLUMN_NAME: self._model.encode(chunk.text), | ||
| "num_tokens": chunk.num_tokens, | ||
| } | ||
| ] | ||
| ) | ||
|
|
||
| def retrieve( | ||
| self, | ||
| prompt: str, | ||
| *, | ||
| chat_id: RagnaId, | ||
| chunk_size: int = 500, | ||
| num_tokens: int = 1024, | ||
| ) -> list[Source]: | ||
| table = self._db.open_table(str(chat_id)) | ||
|
|
||
| # We cannot retrieve source by a maximum number of tokens. Thus, we estimate how | ||
| # many sources we have to query. We overestimate by a factor of two to avoid | ||
| # retrieving to few sources and needed to query again. | ||
| limit = int(num_tokens * 2 / chunk_size) | ||
| results = table.search().limit(limit).to_arrow() | ||
|
|
||
| return list( | ||
| take_sources_up_to_max_tokens( | ||
| ( | ||
| Source( | ||
| id=RagnaId.make(), | ||
| document_id=RagnaId(result["document_id"]), | ||
| document_name=result["document_name"], | ||
| location=result["page_numbers"], | ||
| content=result["text"], | ||
| num_tokens=result["num_tokens"], | ||
| ) | ||
| for result in results.to_pylist() | ||
| ), | ||
| max_tokens=num_tokens, | ||
| ) | ||
| ) |