feat: add hybrid search to node and rust SDKs (#1940)

Support hybrid search in both rust and node SDKs.

- Adds a new rerankers package to rust LanceDB, with the implementation
of the default RRF reranker
- Adds a new hybrid package to lancedb, with some helper methods related
to hybrid search such as normalizing scores and converting score column
to rank columns
- Adds capability to LanceDB VectorQuery to perform hybrid search if it
has both a nearest vector and full text search parameters.
- Adds wrappers for reranker implementations to nodejs SDK.

Additional rerankers will be added in followup PRs

https://github.com/lancedb/lancedb/issues/1921

---
Notes about how the rust rerankers are wrapped for calling from JS:

I wanted to keep the core reranker logic, and the invocation of the
reranker by the query code, in Rust. This aligns with the philosophy of
the new node SDK where it's just a thin wrapper around Rust. However, I
also wanted to have support for users who want to add custom rerankers
written in Javascript.

When we add a reranker to the query from Javascript, it adds a special
Rust reranker that has a callback to the Javascript code (which could
then turn around and call an underlying Rust reranker implementation if
desired). This adds a bit of complexity, but overall I think it moves us
in the right direction of having the majority of the query logic in the
underlying Rust SDK while keeping the option open to support custom
Javascript Rerankers.
This commit is contained in:
Bert
2024-12-30 09:03:41 -05:00
committed by GitHub
parent 0cb6da6b7e
commit c9f248b058
16 changed files with 1295 additions and 6 deletions

View File

@@ -12,7 +12,10 @@ categories.workspace = true
crate-type = ["cdylib"]
[dependencies]
async-trait.workspace = true
arrow-ipc.workspace = true
arrow-array.workspace = true
arrow-schema.workspace = true
env_logger.workspace = true
futures.workspace = true
lancedb = { path = "../rust/lancedb", features = ["remote"] }

View File

@@ -20,6 +20,8 @@ import * as arrow18 from "apache-arrow-18";
import {
convertToTable,
fromBufferToRecordBatch,
fromRecordBatchToBuffer,
fromTableToBuffer,
makeArrowTable,
makeEmptyTable,
@@ -553,5 +555,28 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
});
});
});
describe("converting record batches to buffers", function () {
it("can convert to buffered record batch and back again", async function () {
const records = [
{ text: "dog", vector: [0.1, 0.2] },
{ text: "cat", vector: [0.3, 0.4] },
];
const table = await convertToTable(records);
const batch = table.batches[0];
const buffer = await fromRecordBatchToBuffer(batch);
const result = await fromBufferToRecordBatch(buffer);
expect(JSON.stringify(batch.toArray())).toEqual(
JSON.stringify(result?.toArray()),
);
});
it("converting from buffer returns null if buffer has no record batches", async function () {
const result = await fromBufferToRecordBatch(Buffer.from([0x01, 0x02])); // bad data
expect(result).toEqual(null);
});
});
},
);

View File

@@ -0,0 +1,79 @@
// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright The LanceDB Authors
import { RecordBatch } from "apache-arrow";
import * as tmp from "tmp";
import { Connection, Index, Table, connect, makeArrowTable } from "../lancedb";
import { RRFReranker } from "../lancedb/rerankers";
describe("rerankers", function () {
let tmpDir: tmp.DirResult;
let conn: Connection;
let table: Table;
beforeEach(async () => {
tmpDir = tmp.dirSync({ unsafeCleanup: true });
conn = await connect(tmpDir.name);
table = await conn.createTable("mytable", [
{ vector: [0.1, 0.1], text: "dog" },
{ vector: [0.2, 0.2], text: "cat" },
]);
await table.createIndex("text", {
config: Index.fts(),
replace: true,
});
});
it("will query with the custom reranker", async function () {
const expectedResult = [
{
text: "albert",
// biome-ignore lint/style/useNamingConvention: this is the lance field name
_relevance_score: 0.99,
},
];
class MyCustomReranker {
async rerankHybrid(
_query: string,
_vecResults: RecordBatch,
_ftsResults: RecordBatch,
): Promise<RecordBatch> {
// no reranker logic, just return some static data
const table = makeArrowTable(expectedResult);
return table.batches[0];
}
}
let result = await table
.query()
.nearestTo([0.1, 0.1])
.fullTextSearch("dog")
.rerank(new MyCustomReranker())
.select(["text"])
.limit(5)
.toArray();
result = JSON.parse(JSON.stringify(result)); // convert StructRow to Object
expect(result).toEqual([
{
text: "albert",
// biome-ignore lint/style/useNamingConvention: this is the lance field name
_relevance_score: 0.99,
},
]);
});
it("will query with RRFReranker", async function () {
// smoke test to see if the Rust wrapping Typescript is wired up correctly
const result = await table
.query()
.nearestTo([0.1, 0.1])
.fullTextSearch("dog")
.rerank(await RRFReranker.create())
.select(["text"])
.limit(5)
.toArray();
expect(result).toHaveLength(2);
});
});

View File

@@ -27,7 +27,9 @@ import {
List,
Null,
RecordBatch,
RecordBatchFileReader,
RecordBatchFileWriter,
RecordBatchReader,
RecordBatchStreamWriter,
Schema,
Struct,
@@ -810,6 +812,30 @@ export async function fromDataToBuffer(
}
}
/**
* Read a single record batch from a buffer.
*
* Returns null if the buffer does not contain a record batch
*/
export async function fromBufferToRecordBatch(
data: Buffer,
): Promise<RecordBatch | null> {
const iter = await RecordBatchFileReader.readAll(Buffer.from(data)).next()
.value;
const recordBatch = iter?.next().value;
return recordBatch || null;
}
/**
* Create a buffer containing a single record batch
*/
export async function fromRecordBatchToBuffer(
batch: RecordBatch,
): Promise<Buffer> {
const writer = new RecordBatchFileWriter().writeAll([batch]);
return Buffer.from(await writer.toUint8Array());
}
/**
* Serialize an Arrow Table into a buffer using the Arrow IPC Stream serialization
*

View File

@@ -62,6 +62,7 @@ export { Index, IndexOptions, IvfPqOptions } from "./indices";
export { Table, AddDataOptions, UpdateOptions, OptimizeOptions } from "./table";
export * as embedding from "./embedding";
export * as rerankers from "./rerankers";
/**
* Connect to a LanceDB instance at the given URI.

View File

@@ -16,6 +16,8 @@ import {
Table as ArrowTable,
type IntoVector,
RecordBatch,
fromBufferToRecordBatch,
fromRecordBatchToBuffer,
tableFromIPC,
} from "./arrow";
import { type IvfPqOptions } from "./indices";
@@ -25,6 +27,7 @@ import {
Table as NativeTable,
VectorQuery as NativeVectorQuery,
} from "./native";
import { Reranker } from "./rerankers";
export class RecordBatchIterator implements AsyncIterator<RecordBatch> {
private promisedInner?: Promise<NativeBatchIterator>;
private inner?: NativeBatchIterator;
@@ -542,6 +545,27 @@ export class VectorQuery extends QueryBase<NativeVectorQuery> {
return this;
}
}
rerank(reranker: Reranker): VectorQuery {
super.doCall((inner) =>
inner.rerank({
rerankHybrid: async (_, args) => {
const vecResults = await fromBufferToRecordBatch(args.vecResults);
const ftsResults = await fromBufferToRecordBatch(args.ftsResults);
const result = await reranker.rerankHybrid(
args.query,
vecResults as RecordBatch,
ftsResults as RecordBatch,
);
const buffer = fromRecordBatchToBuffer(result);
return buffer;
},
}),
);
return this;
}
}
/** A builder for LanceDB queries. */

View File

@@ -0,0 +1,17 @@
// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright The LanceDB Authors
import { RecordBatch } from "apache-arrow";
export * from "./rrf";
// Interface for a reranker. A reranker is used to rerank the results from a
// vector and FTS search. This is useful for combining the results from both
// search methods.
export interface Reranker {
rerankHybrid(
query: string,
vecResults: RecordBatch,
ftsResults: RecordBatch,
): Promise<RecordBatch>;
}

View File

@@ -0,0 +1,40 @@
// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright The LanceDB Authors
import { RecordBatch } from "apache-arrow";
import { fromBufferToRecordBatch, fromRecordBatchToBuffer } from "../arrow";
import { RrfReranker as NativeRRFReranker } from "../native";
/**
* Reranks the results using the Reciprocal Rank Fusion (RRF) algorithm.
*
* Internally this uses the Rust implementation
*/
export class RRFReranker {
private inner: NativeRRFReranker;
constructor(inner: NativeRRFReranker) {
this.inner = inner;
}
public static async create(k: number = 60) {
return new RRFReranker(
await NativeRRFReranker.tryNew(new Float32Array([k])),
);
}
async rerankHybrid(
query: string,
vecResults: RecordBatch,
ftsResults: RecordBatch,
): Promise<RecordBatch> {
const buffer = await this.inner.rerankHybrid(
query,
await fromRecordBatchToBuffer(vecResults),
await fromRecordBatchToBuffer(ftsResults),
);
const recordBatch = await fromBufferToRecordBatch(buffer);
return recordBatch as RecordBatch;
}
}

View File

@@ -24,6 +24,7 @@ mod iterator;
pub mod merge;
mod query;
pub mod remote;
mod rerankers;
mod table;
mod util;

View File

@@ -12,6 +12,8 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::sync::Arc;
use lancedb::index::scalar::FullTextSearchQuery;
use lancedb::query::ExecutableQuery;
use lancedb::query::Query as LanceDbQuery;
@@ -25,6 +27,8 @@ use napi_derive::napi;
use crate::error::convert_error;
use crate::error::NapiErrorExt;
use crate::iterator::RecordBatchIterator;
use crate::rerankers::Reranker;
use crate::rerankers::RerankerCallbacks;
use crate::util::parse_distance_type;
#[napi]
@@ -218,6 +222,14 @@ impl VectorQuery {
self.inner = self.inner.clone().with_row_id();
}
#[napi]
pub fn rerank(&mut self, callbacks: RerankerCallbacks) {
self.inner = self
.inner
.clone()
.rerank(Arc::new(Reranker::new(callbacks)));
}
#[napi(catch_unwind)]
pub async fn execute(
&self,

147
nodejs/src/rerankers.rs Normal file
View File

@@ -0,0 +1,147 @@
// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright The LanceDB Authors
use arrow_array::RecordBatch;
use async_trait::async_trait;
use napi::{
bindgen_prelude::*,
threadsafe_function::{ErrorStrategy, ThreadsafeFunction},
};
use napi_derive::napi;
use lancedb::ipc::batches_to_ipc_file;
use lancedb::rerankers::Reranker as LanceDBReranker;
use lancedb::{error::Error, ipc::ipc_file_to_batches};
use crate::error::NapiErrorExt;
/// Reranker implementation that "wraps" a NodeJS Reranker implementation.
/// This contains references to the callbacks that can be used to invoke the
/// reranking methods on the NodeJS implementation and handles serializing the
/// record batches to Arrow IPC buffers.
#[napi]
pub struct Reranker {
/// callback to the Javascript which will call the rerankHybrid method of
/// some Reranker implementation
rerank_hybrid: ThreadsafeFunction<RerankHybridCallbackArgs, ErrorStrategy::CalleeHandled>,
}
#[napi]
impl Reranker {
#[napi]
pub fn new(callbacks: RerankerCallbacks) -> Self {
let rerank_hybrid = callbacks
.rerank_hybrid
.create_threadsafe_function(0, move |ctx| Ok(vec![ctx.value]))
.unwrap();
Self { rerank_hybrid }
}
}
#[async_trait]
impl lancedb::rerankers::Reranker for Reranker {
async fn rerank_hybrid(
&self,
query: &str,
vector_results: RecordBatch,
fts_results: RecordBatch,
) -> lancedb::error::Result<RecordBatch> {
let callback_args = RerankHybridCallbackArgs {
query: query.to_string(),
vec_results: batches_to_ipc_file(&[vector_results])?,
fts_results: batches_to_ipc_file(&[fts_results])?,
};
let promised_buffer: Promise<Buffer> = self
.rerank_hybrid
.call_async(Ok(callback_args))
.await
.map_err(|e| Error::Runtime {
message: format!("napi error status={}, reason={}", e.status, e.reason),
})?;
let buffer = promised_buffer.await.map_err(|e| Error::Runtime {
message: format!("napi error status={}, reason={}", e.status, e.reason),
})?;
let mut reader = ipc_file_to_batches(buffer.to_vec())?;
let result = reader.next().ok_or(Error::Runtime {
message: "reranker result deserialization failed".to_string(),
})??;
return Ok(result);
}
}
impl std::fmt::Debug for Reranker {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.write_str("NodeJSRerankerWrapper")
}
}
#[napi(object)]
pub struct RerankerCallbacks {
pub rerank_hybrid: JsFunction,
}
#[napi(object)]
pub struct RerankHybridCallbackArgs {
pub query: String,
pub vec_results: Vec<u8>,
pub fts_results: Vec<u8>,
}
fn buffer_to_record_batch(buffer: Buffer) -> Result<RecordBatch> {
let mut reader = ipc_file_to_batches(buffer.to_vec()).default_error()?;
reader
.next()
.ok_or(Error::InvalidInput {
message: "expected buffer containing record batch".to_string(),
})
.default_error()?
.map_err(Error::from)
.default_error()
}
/// Wrapper around rust RRFReranker
#[napi]
pub struct RRFReranker {
inner: lancedb::rerankers::rrf::RRFReranker,
}
#[napi]
impl RRFReranker {
#[napi]
pub async fn try_new(k: &[f32]) -> Result<Self> {
let k = k
.first()
.copied()
.ok_or(Error::InvalidInput {
message: "must supply RRF Reranker constructor arg 'k'".to_string(),
})
.default_error()?;
Ok(Self {
inner: lancedb::rerankers::rrf::RRFReranker::new(k),
})
}
#[napi]
pub async fn rerank_hybrid(
&self,
query: String,
vec_results: Buffer,
fts_results: Buffer,
) -> Result<Buffer> {
let vec_results = buffer_to_record_batch(vec_results)?;
let fts_results = buffer_to_record_batch(fts_results)?;
let result = self
.inner
.rerank_hybrid(&query, vec_results, fts_results)
.await
.unwrap();
let result_buff = batches_to_ipc_file(&[result]).default_error()?;
Ok(Buffer::from(result_buff.as_ref()))
}
}