Lance is a columnar data format that is easy and fast to version, query and train on. It’s designed to be used with images, videos, 3D point clouds, audio and of course tabular data. The key features of Lance include:

• High-performance random access: 100x faster than Parquet.

• Vector search: find nearest neighbors in under 1 millisecond and combine OLAP-queries with vector search.

• Zero-copy, automatic versioning: manage versions of your data automatically, and reduce redundancy with zero-copy logic built-in.

• Ecosystem integrations: Apache-Arrow, DuckDB and more on the way.

Installation

pip install pylance

Converting to Lance

import lance

import pandas as pd
import pyarrow as pa
import pyarrow.dataset

df = pd.DataFrame({"a": [5], "b": [10]})
uri = "/tmp/test.parquet"
tbl = pa.Table.from_pandas(df)
pa.dataset.write_dataset(tbl, uri, format='parquet')

parquet = pa.dataset.dataset(uri, format='parquet')
lance.write_dataset(parquet, "/tmp/test.lance")

Reading Lance data

dataset = lance.dataset("/tmp/test.lance")
assert isinstance(dataset, pa.dataset.Dataset)

Pandas

df = dataset.to_table().to_pandas()
df

DuckDB

import duckdb

tbl = dataset.to_table()  # next release of duckdb will have pushdowns enabled
duckdb.query("SELECT * FROM tbl LIMIT 10").to_df()

Vector search

Download an indexed sift dataset, and unzip it into vec_data.lance

wget https://eto-public.s3.us-west-2.amazonaws.com/datasets/sift/sift_ivf256_pq16.tar.gz
tar -xzf sift_ivf256_pq16.tar.gz

# Get top 10 similar vectors
import lance
import duckdb
import numpy as np

uri = "vec_data.lance"
dataset = lance.dataset(uri)

# Sample 100 query vectors
tbl = dataset.to_table()
sample = duckdb.query("SELECT vector FROM tbl USING SAMPLE 100").to_df()
query_vectors = np.array([np.array(x) for x in sample.vector])

# Get nearest neighbors for all of them
rs = [dataset.to_table(nearest={"column": "vector", 
                                "k": 10, 
                                "q": query_vectors[i, :]}) 
      for i in range(query_vectors.shape[0])]

Here we will highlight a few aspects of Lance’s design. For more details, see the full Lance design document.

Vector index: Vector index for similarity search over embedding space

Encodings: to achieve both fast columnar scan and sub-linear point queries, Lance uses custom encodings and layouts.

Nested fields: Lance stores each subfield as a separate column to support efficient filters like “find images where detected objects include cats”.

Versioning: a Manifest can be used to record snapshots. Currently we support creating new versions automatically via appends, overwrites, and index creation

Fast updates (ROADMAP): Updates will be supported via write-ahead logs.

Rich secondary indices (ROADMAP):

• Inverted index for fuzzy search over many label / annotation fields

We used the sift dataset to benchmark our results with 1M vectors of 128D

1. For 100 randomly sampled query vectors, we get <1ms average response time (on a 2023 m2 macbook air)

2. ANN is always a trade-off between recall and performance

We create a Lance dataset using the Oxford Pet dataset to do some preliminary performance testing of Lance as compared to Parquet and raw image/xmls. For analytics queries, Lance is 50-100x better than reading the raw metadata. For batched random access, Lance is 100x better than both parquet and raw files.

Machine Learning development cycle involves the steps:

graph LR
    A[Collection] --> B[Exploration];
    B --> C[Analytics];
    C --> D[Feature Engineer];
    D --> E[Training];
    E --> F[Evaluation];
    F --> C;
    E --> G[Deployment];
    G --> H[Monitoring];
    H --> A;

People use different data representations to varying stages for the performance or limited by the tooling available. The academia mainly uses XML / JSON for annotations and zipped images/sensors data for deep learning, which is difficult to integrated into data infrastructure and slow to train over cloud storage. While the industry uses data lake (Parquet-based techniques, i.e., Delta Lake, Iceberg) or data warehouse (AWS Redshift or Google BigQuery) to collect and analyze data, they have to convert the data into training-friendly formats, such as Rikai/Petastorm or Tfrecord. Multiple single-purpose data transforms, as well as syncing copies between cloud storage to local training instances have become a common practice among ML practices.

While each of the existing data formats excel at its original designed workload, we need a new data format to tailored for multistage ML development cycle to reduce the fraction in tools and data silos.

A comparison of different data formats in each stage of ML development cycle.

• Lance: A New Columnar Data Format, Scipy 2022, Austin, TX. July, 2022.