JinHai-CN / pymilvus

Python client for Milvus distributed high-performance vector database system.

Milvus Python SDK

Using Milvus python sdk for Milvus

Download

Pymilvus only supports python >= 3.4, is fully tested under 3.4, 3.5, 3.6.

Python 3.7 can work, but not fully tested yet.

Pymilvus can be downloaded using pip. If no use, try pip3

$ pip install pymilvus

Upgrade to newest version

$ pip install --upgrade pymilvus

Import

from milvus import Milvus, Prepare, IndexType, Status

Getting started

Initial a Milvus instance and connect to the sever

>>> milvus = Milvus()

>>> milvus.connect(host='SERVER-HOST', port='SERVER-PORT')
Status(code=0, message="Success")

Once successfully connected, you can get the version of server

>>> milvus.server_version()
0.0.0  # this is example version, the real version may vary

---

Add a new table

First using Prepare to create param

>>> param = Prepare.table_schema(table_name='test01', dimension=256, index_type=IndexType.IDMAP,
                                    store_raw_vector=False)

Then create table

>>> milvus.create_table(param)
Status(message='Table test01 created!', code=0)

Describe the table we just created

>>> milvus.describe_table('test01')
(Status(code=0, message='Success!'), TableSchema(table_name='test01',dimension=256, index_type=1, store_raw_vector=False))

---

Add vectors into table test01

First Prepare binary vectors of 256-dimension.

• Note that random and pprint we used here is for creating fake vectors data and pretty print, you may not need them in your project

>>> import random
>>> from pprint import pprint

>>> dim = 256  # Dimension of the vector

# Initialize 20 vectors of 256-dimension
>>> fake_vectors = [[random.random() for _ in range(dim)] for _ in range(20)]
>>> vectors = Prepare.records(fake_vectors)  # This will transfer fake_vector to binary data

Then add vectors into table test01

>>> status, ids = milvus.add_vectors(table_name='test01', records=vectors)
>>> print(status)
Status(code=0, message='Success')
>>> pprint(ids) # List of ids returned
23455321135511233
12245748929023489
...

---

Search vectors

# prepare 5 vectors of 256-dimension
>>> q_records = Prepare.records([random.random() for _ in range(dim)] for _ in range(5)]

Then get results

>>> status, results = milvus.search_vectors(table_name='test01', query_records=q_records, top_k=10)
>>> print(status)
Status(code=0, message='Success')
>>> pprint(results) # Searched top_k vectors

Disconnect with the server

>>> milvus.disconnect()
Status(code=0, message='Success')

---

Example python

There are some small examples in examples/, you can find more guide there.

Build docs

$ sphinx-build -b html doc/en/ doc/en/build

If you encounter any problems or bugs, please add new issues