rust/lancedb
1
0
Fork 0
mirror of https://github.com/lancedb/lancedb.git synced 2026-05-26 00:10:39 +00:00
Code Issues Packages Projects Releases Wiki Activity

docs: add the async python API to the docs (#1156)

Browse Source
This commit is contained in:
Weston Pace
2024-03-26 07:54:16 -05:00
committed by GitHub
parent 98c1e635b3
commit c37a28abbd
13 changed files with 624 additions and 400 deletions
Download Patch File Download Diff File Expand all files Collapse all files

82
docs/src/basic.md
View File

@@ -48,11 +48,20 @@
=== "Python"
```python
import lancedb
uri = "data/sample-lancedb"
db = lancedb.connect(uri)
```
```python
--8<-- "python/python/tests/docs/test_basic.py:imports"
--8<-- "python/python/tests/docs/test_basic.py:connect"
--8<-- "python/python/tests/docs/test_basic.py:connect_async"
```
!!! note "Asynchronous Python API"
The asynchronous Python API is new and has some slight differences compared
to the synchronous API. Feel free to start using the asynchronous version.
Once all features have migrated we will start to move the synchronous API to
use the same syntax as the asynchronous API. To help with this migration we
have created a [migration guide](migration.md) detailing the differences.
=== "Typescript"
@@ -82,15 +91,14 @@ If you need a reminder of the uri, you can call `db.uri()`.
### Create a table from initial data
If you have data to insert into the table at creation time, you can simultaneously create a
table and insert the data into it. The schema of the data will be used as the schema of the
table and insert the data into it. The schema of the data will be used as the schema of the
table.
=== "Python"
```python
tbl = db.create_table("my_table",
data=[{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
{"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
--8<-- "python/python/tests/docs/test_basic.py:create_table"
--8<-- "python/python/tests/docs/test_basic.py:create_table_async"
```
If the table already exists, LanceDB will raise an error by default.
@@ -100,10 +108,8 @@ table.
You can also pass in a pandas DataFrame directly:
```python
import pandas as pd
df = pd.DataFrame([{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
{"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
tbl = db.create_table("table_from_df", data=df)
--8<-- "python/python/tests/docs/test_basic.py:create_table_pandas"
--8<-- "python/python/tests/docs/test_basic.py:create_table_async_pandas"
```
=== "Typescript"
@@ -138,15 +144,14 @@ table.
Sometimes you may not have the data to insert into the table at creation time.
In this case, you can create an empty table and specify the schema, so that you can add
data to the table at a later time (as long as it conforms to the schema). This is
data to the table at a later time (as long as it conforms to the schema). This is
similar to a `CREATE TABLE` statement in SQL.
=== "Python"
```python
import pyarrow as pa
schema = pa.schema([pa.field("vector", pa.list_(pa.float32(), list_size=2))])
tbl = db.create_table("empty_table", schema=schema)
--8<-- "python/python/tests/docs/test_basic.py:create_empty_table"
--8<-- "python/python/tests/docs/test_basic.py:create_empty_table_async"
```
=== "Typescript"
@@ -168,7 +173,8 @@ Once created, you can open a table as follows:
=== "Python"
```python
tbl = db.open_table("my_table")
--8<-- "python/python/tests/docs/test_basic.py:open_table"
--8<-- "python/python/tests/docs/test_basic.py:open_table_async"
```
=== "Typescript"
@@ -188,7 +194,8 @@ If you forget the name of your table, you can always get a listing of all table
=== "Python"
```python
print(db.table_names())
--8<-- "python/python/tests/docs/test_basic.py:table_names"
--8<-- "python/python/tests/docs/test_basic.py:table_names_async"
```
=== "Javascript"
@@ -210,15 +217,8 @@ After a table has been created, you can always add more data to it as follows:
=== "Python"
```python
# Option 1: Add a list of dicts to a table
data = [{"vector": [1.3, 1.4], "item": "fizz", "price": 100.0},
{"vector": [9.5, 56.2], "item": "buzz", "price": 200.0}]
tbl.add(data)
# Option 2: Add a pandas DataFrame to a table
df = pd.DataFrame(data)
tbl.add(data)
--8<-- "python/python/tests/docs/test_basic.py:add_data"
--8<-- "python/python/tests/docs/test_basic.py:add_data_async"
```
=== "Typescript"
@@ -240,7 +240,8 @@ Once you've embedded the query, you can find its nearest neighbors as follows:
=== "Python"
```python
tbl.search([100, 100]).limit(2).to_pandas()
--8<-- "python/python/tests/docs/test_basic.py:vector_search"
--8<-- "python/python/tests/docs/test_basic.py:vector_search_async"
```
This returns a pandas DataFrame with the results.
@@ -274,7 +275,8 @@ LanceDB allows you to create an ANN index on a table as follows:
=== "Python"
```py
tbl.create_index()
--8<-- "python/python/tests/docs/test_basic.py:create_index"
--8<-- "python/python/tests/docs/test_basic.py:create_index_async"
```
=== "Typescript"
@@ -286,15 +288,15 @@ LanceDB allows you to create an ANN index on a table as follows:
=== "Rust"
```rust
--8<-- "rust/lancedb/examples/simple.rs:create_index"
--8<-- "rust/lancedb/examples/simple.rs:create_index"
```
!!! note "Why do I need to create an index manually?"
LanceDB does not automatically create the ANN index for two reasons. The first is that it's optimized
for really fast retrievals via a disk-based index, and the second is that data and query workloads can
be very diverse, so there's no one-size-fits-all index configuration. LanceDB provides many parameters
to fine-tune index size, query latency and accuracy. See the section on
[ANN indexes](ann_indexes.md) for more details.
LanceDB does not automatically create the ANN index for two reasons. The first is that it's optimized
for really fast retrievals via a disk-based index, and the second is that data and query workloads can
be very diverse, so there's no one-size-fits-all index configuration. LanceDB provides many parameters
to fine-tune index size, query latency and accuracy. See the section on
[ANN indexes](ann_indexes.md) for more details.
## Delete rows from a table
@@ -305,7 +307,8 @@ This can delete any number of rows that match the filter.
=== "Python"
```python
tbl.delete('item = "fizz"')
--8<-- "python/python/tests/docs/test_basic.py:delete_rows"
--8<-- "python/python/tests/docs/test_basic.py:delete_rows_async"
```
=== "Typescript"
@@ -322,7 +325,7 @@ This can delete any number of rows that match the filter.
The deletion predicate is a SQL expression that supports the same expressions
as the `where()` clause (`only_if()` in Rust) on a search. They can be as
simple or complex as needed. To see what expressions are supported, see the
simple or complex as needed. To see what expressions are supported, see the
[SQL filters](sql.md) section.
=== "Python"
@@ -344,7 +347,8 @@ Use the `drop_table()` method on the database to remove a table.
=== "Python"
```python
db.drop_table("my_table")
--8<-- "python/python/tests/docs/test_basic.py:drop_table"
--8<-- "python/python/tests/docs/test_basic.py:drop_table_async"
```
This permanently removes the table and is not recoverable, unlike deleting rows.

76
docs/src/migration.md Normal file
View File

@@ -0,0 +1,76 @@
# Rust-backed Client Migration Guide
In an effort to ensure all clients have the same set of capabilities we have begun migrating the
python and node clients onto a common Rust base library. In python, this new client is part of
the same lancedb package, exposed as an asynchronous client. Once the asynchronous client has
reached full functionality we will begin migrating the synchronous library to be a thin wrapper
around the asynchronous client.
This guide describes the differences between the two APIs and will hopefully assist users
that would like to migrate to the new API.
## Closeable Connections
The Connection now has a `close` method. You can call this when
you are done with the connection to eagerly free resources. Currently
this is limited to freeing/closing the HTTP connection for remote
connections. In the future we may add caching or other resources to
native connections so this is probably a good practice even if you
aren't using remote connections.
In addition, the connection can be used as a context manager which may
be a more convenient way to ensure the connection is closed.
```python
import lancedb
async def my_async_fn():
with await lancedb.connect_async("my_uri") as db:
print(await db.table_names())
```
It is not mandatory to call the `close` method. If you do not call it
then the connection will be closed when the object is garbage collected.
## Closeable Table
The Table now also has a `close` method, similar to the connection. This
can be used to eagerly free the cache used by a Table object. Similar to
the connection, it can be used as a context manager and it is not mandatory
to call the `close` method.
### Changes to Table APIs
- Previously `Table.schema` was a property. Now it is an async method.
- The method `Table.__len__` was removed and `len(table)` will no longer
work. Use `Table.count_rows` instead.
### Creating Indices
The `Table.create_index` method is now used for creating both vector indices
and scalar indices. It currently requires a column name to be specified (the
column to index). Vector index defaults are now smarter and scale better with
the size of the data.
To specify index configuration details you will need to specify which kind of
index you are using.
### Querying
The `Table.search` method has been renamed to `AsyncTable.vector_search` for
clarity.
## Features not yet supported
The following features are not yet supported by the asynchronous API. However,
we plan to support them soon.
- You cannot specify an embedding function when creating or opening a table.
You must calculate embeddings yourself if using the asynchronous API
- The merge insert operation is not supported in the asynchronous API
- Cleanup / compact / optimize indices are not supported in the asynchronous API
- add / alter columns is not supported in the asynchronous API
- The asynchronous API does not yet support any full text search or reranking
search
- Remote connections to LanceDb Cloud are not yet supported.
- The method Table.head is not yet supported.

49
docs/src/python/python.md
View File

@@ -8,17 +8,20 @@ This section contains the API reference for the OSS Python API.
pip install lancedb
```
## Connection
The following methods describe the synchronous API client. There
is also an [asynchronous API client](#connections-asynchronous).
## Connections (Synchronous)
::: lancedb.connect
::: lancedb.db.DBConnection
## Table
## Tables (Synchronous)
::: lancedb.table.Table
## Querying
## Querying (Synchronous)
::: lancedb.query.Query
@@ -86,4 +89,42 @@ pip install lancedb
::: lancedb.rerankers.cross_encoder.CrossEncoderReranker
::: lancedb.rerankers.openai.OpenaiReranker
::: lancedb.rerankers.openai.OpenaiReranker
## Connections (Asynchronous)
Connections represent a connection to a LanceDb database and
can be used to create, list, or open tables.
::: lancedb.connect_async
::: lancedb.db.AsyncConnection
## Tables (Asynchronous)
Table hold your actual data as a collection of records / rows.
::: lancedb.table.AsyncTable
## Indices (Asynchronous)
Indices can be created on a table to speed up queries. This section
lists the indices that LanceDb supports.
::: lancedb.index.BTree
::: lancedb.index.IvfPq
## Querying (Asynchronous)
Queries allow you to return data from your database. Basic queries can be
created with the [AsyncTable.query][lancedb.table.AsyncTable.query] method
to return the entire (typically filtered) table. Vector searches return the
rows nearest to a query vector and can be created with the
[AsyncTable.vector_search][lancedb.table.AsyncTable.vector_search] method.
::: lancedb.query.AsyncQueryBase
::: lancedb.query.AsyncQuery
::: lancedb.query.AsyncVectorQuery