# DashVector Client Python Library
DashVector is a scalable and fully-managed vector-database service for building various machine learning applications. The DashVector client SDK is your gateway to access the DashVector service.
For more information about DashVector, please visit: https://help.aliyun.com/document_detail/2510225.html
## Installation
To install the DashVector client Python SDK, simply run:
```shell
pip install dashvector
```
## QuickStart
```python
import numpy as np
import dashvector
# Use DashVector `Client` api to communicate with the backend vectorDB service.
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
# Create a collection named "quickstart" with dimension of 4, using the default Cosine distance metric
rsp = client.create(name='quickstart', dimension=4)
assert rsp
# Get a collection by name
collection = client.get(name='quickstart')
# Operations on 'Collection' includes Inert/Query/Upsert/Update/Delete/Fetch of docs
# Here we insert sample data (4-dimensional vectors) in batches of 16
collection.insert(
[
dashvector.Doc(id=str(i), vector=np.random.rand(4), fields={'anykey': 'anyvalue'})
for i in range(16)
]
)
# Query a vector from the collection
docs = collection.query([0.1, 0.2, 0.3, 0.4], topk=5)
print(docs)
# Get statistics about collection
stats = collection.stats()
print(stats)
# Delete a collection by name
client.delete(name='quickstart')
```
## Reference
### Create a Client
`Client` host various APIs for interacting with DashVector `Collection`.
```python
dashvector.Client(
api_key: str,
endpoint: str = 'dashvector.cn-hangzhou.aliyuncs.com',
protocal: dashvector.DashVectorProtocol = dashvector.DashVectorProtocol.GRPC,
timeout: float = 10.0
) -> Client
```
| Parameters | Type | Required | Description |
|------------|--------------------|----------|----------------------------------------------------------------------------------------------|
| api_key | str | Yes | Your DashVector API-KEY |
| endpoint | str | No | Service Endpoint. <br/>Default value: `dashvector.cn-hangzhou.aliyuncs.com` |
| protocol | DashVectorProtocol | No | Communication protocol, support HTTP and GRPC. <br/>Default value: `DashVectorProtocol.GRPC` |
| timeout | float | No | Timeout period (in seconds), -1 means no timeout. <br/>Default value: `10.0` |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
assert client
```
### Create Collection
```python
Client.create(
name: str,
dimension: int,
dtype: Union[Type[int], Type[float]] = float,
fields_schema: Optional[Dict[str, Union[Type[str], Type[int], Type[float], Type[bool]]]] = None,
metric: str = 'cosine',
timeout: Optional[int] = None
) -> DashVectorResponse
```
| Parameters | Type | Required | Description |
|----------------|----------------------------------------------------------------------------|----------|------------------------------------------------------------------------------------------------------------------|
| name | str | Yes | The name of the Collection to create. |
| dimension | int | Yes | The dimensions of the Collection's vectors. Valid values: 1-20,000 |
| dtype | Union[Type[int], Type[float]] | No | The date type of the Collection's vectors.<br/>Default value: `Type[float]` |
| fields_schema | Optional[Dict[str, Union[Type[str], Type[int], Type[float], Type[bool]]]] | No | Fields schema of the Collection.<br/>Default value: `None`<br/>e.g. `{"name": str, "age": int}` |
| metric | str | No | Vector similarity metric. For `cosine`, dtype must be `float`.<br/>Valid values:<br/> 1. (Default)`cosine`<br/>2. `dotproduct`<br/>3. `euclidean` |
| timeout | Optional[int] | No | Timeout period (in seconds), -1 means asynchronous creation collection.<br/>Default value: `None` |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
rsp = client.create('YOUR-COLLECTION-NAME', dimension=4)
assert rsp
```
### List Collections
`Client.list() -> DashVectorResponse`
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collections = client.list()
for collection in collections:
print(collection)
# outputs:
# 'quickstart'
```
### Describe Collection
`Client.describe(name: str) -> DashVectorResponse`
| Parameters | Type | Required | Description |
|------------|-------|----------|-----------------------------------------|
| name | str | Yes | The name of the Collection to describe. |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
rsp = client.describe('YOUR-COLLECTION-NAME')
print(rsp)
# example output:
# {
# "request_id": "8d3ac14e-5382-4736-b77c-4318761ddfab",
# "code": 0,
# "message": "",
# "output": {
# "name": "quickstart",
# "dimension": 4,
# "dtype": "FLOAT",
# "metric": "dotproduct",
# "fields_schema": {
# "name": "STRING",
# "age": "INT",
# "height": "FLOAT"
# },
# "status": "SERVING",
# "partitions": {
# "default": "SERVING"
# }
# }
# }
```
### Delete Collection
`Client.delete(name: str) -> DashVectorResponse`
| Parameters | Type | Required | Description |
|------------|-------|----------|---------------------------------------|
| name | str | Yes | The name of the Collection to delete. |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
client.delete('YOUR-COLLECTION-NAME')
```
### Get a Collection Instance
`Collection` provides APIs for accessing `Doc` and `Partition`
`Client.get(name: str) -> Collection`
| Parameters | Type | Required | Description |
|------------|-------|----------|------------------------------------|
| name | str | Yes | The name of the Collection to get. |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
assert collection
```
### Describe Collection Statistics
`Collection.stats() -> DashVectorResponse`
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
rsp = collection.stats()
print(rsp)
# example output:
# {
# "request_id": "14448bcb-c9a3-49a8-9152-0de3990bce59",
# "code": 0,
# "message": "Success",
# "output": {
# "total_doc_count": "26",
# "index_completeness": 1.0,
# "partitions": {
# "default": {
# "total_doc_count": "26"
# }
# }
# }
# }
```
### Insert/Update/Upsert Docs
```python
Collection.insert(
docs: Union[Doc, List[Doc], Tuple, List[Tuple]],
partition: Optional[str] = None,
async_req: False
) -> DashVectorResponse
```
| Parameters | Type | Required | Description |
|------------|-------------------------------------------|----------|------------------------------------------------------------------------|
| docs | Union[Doc, List[Doc], Tuple, List[Tuple]] | Yes | The docs to Insert/Update/Upsert. |
| partition | Optional[str] | No | Name of the partition to Insert/Update/Upsert.<br/>Default value: `None` |
| async_req | bool | No | Enable async request or not.<br/>Default value: `False` |
Example:
```python
import dashvector
import numpy as np
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
# insert a doc with Tuple
collection.insert(('YOUR-DOC-ID1', [0.1, 0.2, 0.3, 0.4]))
collection.insert(('YOUR-DOC-ID2', [0.2, 0.3, 0.4, 0.5], {'age': 30, 'name': 'alice', 'anykey': 'anyvalue'}))
# insert a doc with dashvector.Doc
collection.insert(
dashvector.Doc(
id='YOUR-DOC-ID3',
vector=[0.3, 0.4, 0.5, 0.6],
fields={'foo': 'bar'}
)
)
# insert in batches
ret = collection.insert(
[
('YOUR-DOC-ID4', [0.2, 0.7, 0.8, 1.3], {'age': 1}),
('YOUR-DOC-ID4', [0.3, 0.6, 0.9, 1.2], {'age': 2}),
('YOUR-DOC-ID6', [0.4, 0.5, 1.0, 1.1], {'age': 3})
]
)
# insert in batches
ret = collection.insert(
[
dashvector.Doc(id=str(i), vector=np.random.rand(4)) for i in range(10)
]
)
# async insert
ret_funture = collection.insert(
[
dashvector.Doc(id=str(i+10), vector=np.random.rand(4)) for i in range(10)
],
async_req=True
)
ret = ret_funture.get()
```
### Query a Collection
```python
Collection.query(
vector: Optional[Union[List[Union[int, float]], np.ndarray]] = None,
id: Optional[str] = None,
topk: int = 10,
filter: Optional[str] = None,
include_vector: bool = False,
partition: Optional[str] = None,
output_fields: Optional[List[str]] = None,
async_req: False
) -> DashVectorResponse
```
| Parameters | Type | Required | Description |
|-----------------|------------------------------------------------------|----------|--------------------------------------------------------------------------------------------------------------|
| vector | Optional[Union[List[Union[int, float]], np.ndarray]] | No | The vector to query |
| id | Optional[str] | No | The doc id to query.<br/>Setting `id` means searching by vector corresponding to the id |
| topk | Optional[str] | No | Number of similarity results to return.<br/>Default value: `10` |
| filter | Optional[str] | No | Expression used to filter results <br/>Default value: None <br/>e.g. `age>20` |
| include_vector | bool | No | Return vector details or not.<br/>Default value: `False` |
| partition | Optional[str] | No | Name of the partition to Query.<br/>Default value: `None` |
| output_fields | Optional[List[str]] | No | List of field names to return.<br/>Default value: `None`, means return all fields<br/>e.g. `['name', 'age']` |
| async_req | bool | No | Enable async request or not.<br/>Default value: `False` |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
match_docs = collection.query([0.1, 0.2, 0.3, 0.4], topk=100, filter='age>20', include_vector=True, output_fields=['age','name','foo'])
if match_docs:
for doc in match_docs:
print(doc.id)
print(doc.vector)
print(doc.fields)
print(doc.score)
```
### Delete Docs
```python
collection.delete(
ids: Union[str, List[str]],
delete_all: bool = False,
partition: Optional[str] = None,
async_req: bool = False
) -> DashVectorResponse
```
| Parameters | Type | Required | Description |
|------------|-----------------------|----------|-----------------------------------------------------------------|
| ids | Union[str, List[str]] | Yes | The id (or list of ids) for the Doc(s) to Delete |
| delete_all | bool | No | Delete all vectors from partition.<br/>Default value: `False` |
| partition | Optional[str] | No | Name of the partition to Delete from.<br/>Default value: `None` |
| async_req | bool | No | Enable async request or not.<br/>Default value: `False` |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
collection.delete(['YOUR-DOC-ID1','YOUR-DOC-ID2'])
```
### Fetch Docs
```python
Collection.fetch(
ids: Union[str, List[str]],
partition: Optional[str] = None,
async_req: bool = False
) -> DashVectorResponse
```
| Parameters | Type | Required | Description |
|------------|-----------------------|----------|----------------------------------------------------------------|
| ids | Union[str, List[str]] | Yes | The id (or list of ids) for the Doc(s) to Fetch |
| partition | Optional[str] | No | Name of the partition to Fetch from.<br/>Default value: `None` |
| async_req | bool | No | Enable async request or not.<br/>Default value: `False` |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
fetch_docs = collection.fetch(['YOUR-DOC-ID1', 'YOUR-DOC-ID2'])
if fetch_docs:
for doc_id in fetch_docs:
doc = fetch_docs[doc_id]
print(doc.id)
print(doc.vector)
print(doc.fields)
```
### Create Collection Partition
`Collection.create_partition(name: str) -> DashVectorResponse`
| Parameters | Type | Required | Description |
|------------|----------------|----------|-------------------------------------------------------------------------------------------------------|
| name | str | Yes | The name of the Partition to Create. |
| timeout | Optional[int] | No | Timeout period (in seconds), -1 means asynchronous creation partition.<br/>Default value: `None` |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
rsp = collection.create_partition('YOUR-PARTITION-NAME')
assert rsp
```
### Delete Collection Partition
`Collection.delete_partition(name: str) -> DashVectorResponse`
| Parameters | Type | Required | Description |
|------------|-------|----------|--------------------------------------|
| name | str | Yes | The name of the Partition to Delete. |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
rsp = collection.delete_partition('YOUR-PARTITION-NAME')
assert rsp
```
### List Collection Partitions
`Collection.list_partitions() -> DashVectorResponse`
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
partitions = collection.list_partitions()
assert partitions
for pt in partitions:
print(pt)
```
### Describe Collection Partition
`Collection.describe_partition(name: str) -> DashVectorResponse`
| Parameters | Type | Required | Description |
|------------|-------|----------|----------------------------------------|
| name | str | Yes | The name of the Partition to Describe. |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
rsp = collection.describe_partition('shoes')
print(rsp)
# example output:
# {"request_id":"296267a7-68e2-483a-87e6-5992d85a5806","code":0,"message":"","output":"SERVING"}
```
### Statistics for Collection Partition
`Collection.stats_partition(name: str) -> DashVectorResponse`
| Parameters | Type | Required | Description |
|------------|-------|----------|----------------------------------------------|
| name | str | Yes | The name of the Partition to get Statistics. |
Example:
```python
import dashvector
client = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')
collection = client.get('YOUR-COLLECTION-NAME')
rsp = collection.stats_partition('shoes')
print(rsp)
# example outptut:
# {
# "code":0,
# "message":"",
# "requests_id":"330a2bcb-e4a7-4fc6-a711-2fe5f8a24e8c",
# "output":{
# "total_doc_count":0
# }
# }
```
## Class
### dashvector.Doc
```python
@dataclass(frozen=True)
class Doc(object):
id: str
vector: Union[List[int], List[float], numpy.ndarray]
fields: Optional[Dict[str, Union[Type[str], Type[int], Type[float], Type[bool]]]] = None
score: float = 0.0
```
### dashvector.DashVectorResponse
```python
class DashVectorResponse(object):
code: DashVectorCode
message: str
request_id: str
output: Any
```
## License
This project is licensed under the Apache License (Version 2.0).
Raw data
{
"_id": null,
"home_page": "https://github.com/alibaba/proxima",
"name": "dashvector",
"maintainer": null,
"docs_url": null,
"requires_python": "<4.0,>=3.7",
"maintainer_email": null,
"keywords": "DashVector, vector, database, cloud",
"author": "Alibaba",
"author_email": null,
"download_url": "https://files.pythonhosted.org/packages/00/c1/2d117f6c52dc1c9226891ce4f1f0e35ba3dd32a4f582b42d71ded2de9e66/dashvector-1.0.16.tar.gz",
"platform": null,
"description": "# DashVector Client Python Library\n\nDashVector is a scalable and fully-managed vector-database service for building various machine learning applications. The DashVector client SDK is your gateway to access the DashVector service.\n\nFor more information about DashVector, please visit: https://help.aliyun.com/document_detail/2510225.html\n\n## Installation\nTo install the DashVector client Python SDK, simply run:\n```shell\npip install dashvector\n```\n\n## QuickStart\n\n```python\nimport numpy as np\nimport dashvector\n\n# Use DashVector `Client` api to communicate with the backend vectorDB service.\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\n\n# Create a collection named \"quickstart\" with dimension of 4, using the default Cosine distance metric\nrsp = client.create(name='quickstart', dimension=4)\nassert rsp\n\n# Get a collection by name\ncollection = client.get(name='quickstart')\n\n# Operations on 'Collection' includes Inert/Query/Upsert/Update/Delete/Fetch of docs\n# Here we insert sample data (4-dimensional vectors) in batches of 16\ncollection.insert(\n [\n dashvector.Doc(id=str(i), vector=np.random.rand(4), fields={'anykey': 'anyvalue'}) \n for i in range(16)\n ]\n)\n\n# Query a vector from the collection\ndocs = collection.query([0.1, 0.2, 0.3, 0.4], topk=5)\nprint(docs)\n\n# Get statistics about collection\nstats = collection.stats()\nprint(stats)\n\n# Delete a collection by name\nclient.delete(name='quickstart')\n```\n\n## Reference\n\n### Create a Client\n`Client` host various APIs for interacting with DashVector `Collection`.\n\n```python\ndashvector.Client(\n api_key: str,\n endpoint: str = 'dashvector.cn-hangzhou.aliyuncs.com',\n protocal: dashvector.DashVectorProtocol = dashvector.DashVectorProtocol.GRPC, \n timeout: float = 10.0\n) -> Client\n```\n\n| Parameters | Type | Required | Description |\n|------------|--------------------|----------|----------------------------------------------------------------------------------------------|\n| api_key | str | Yes | Your DashVector API-KEY |\n| endpoint | str | No | Service Endpoint. <br/>Default value: `dashvector.cn-hangzhou.aliyuncs.com` |\n| protocol | DashVectorProtocol | No | Communication protocol, support HTTP and GRPC. <br/>Default value: `DashVectorProtocol.GRPC` |\n| timeout | float | No | Timeout period (in seconds), -1 means no timeout. <br/>Default value: `10.0` |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\nassert client\n```\n\n### Create Collection\n```python\nClient.create(\n name: str,\n dimension: int,\n dtype: Union[Type[int], Type[float]] = float,\n fields_schema: Optional[Dict[str, Union[Type[str], Type[int], Type[float], Type[bool]]]] = None,\n metric: str = 'cosine',\n timeout: Optional[int] = None\n) -> DashVectorResponse\n```\n\n| Parameters | Type | Required | Description |\n|----------------|----------------------------------------------------------------------------|----------|------------------------------------------------------------------------------------------------------------------|\n| name | str | Yes | The name of the Collection to create. |\n| dimension | int | Yes | The dimensions of the Collection's vectors. Valid values: 1-20,000 |\n| dtype | Union[Type[int], Type[float]] | No | The date type of the Collection's vectors.<br/>Default value: `Type[float]` |\n| fields_schema | Optional[Dict[str, Union[Type[str], Type[int], Type[float], Type[bool]]]] | No | Fields schema of the Collection.<br/>Default value: `None`<br/>e.g. `{\"name\": str, \"age\": int}` |\n| metric | str | No | Vector similarity metric. For `cosine`, dtype must be `float`.<br/>Valid values:<br/> 1. (Default)`cosine`<br/>2. `dotproduct`<br/>3. `euclidean` |\n| timeout | Optional[int] | No | Timeout period (in seconds), -1 means asynchronous creation collection.<br/>Default value: `None` |\n\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\n\nrsp = client.create('YOUR-COLLECTION-NAME', dimension=4)\nassert rsp\n```\n\n### List Collections\n`Client.list() -> DashVectorResponse`\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\n\ncollections = client.list()\n\nfor collection in collections:\n print(collection)\n# outputs:\n# 'quickstart'\n```\n\n### Describe Collection\n`Client.describe(name: str) -> DashVectorResponse`\n\n| Parameters | Type | Required | Description |\n|------------|-------|----------|-----------------------------------------|\n| name | str | Yes | The name of the Collection to describe. |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\nrsp = client.describe('YOUR-COLLECTION-NAME')\n\nprint(rsp)\n# example output:\n# {\n# \"request_id\": \"8d3ac14e-5382-4736-b77c-4318761ddfab\",\n# \"code\": 0,\n# \"message\": \"\",\n# \"output\": {\n# \"name\": \"quickstart\",\n# \"dimension\": 4,\n# \"dtype\": \"FLOAT\",\n# \"metric\": \"dotproduct\",\n# \"fields_schema\": {\n# \"name\": \"STRING\",\n# \"age\": \"INT\",\n# \"height\": \"FLOAT\"\n# },\n# \"status\": \"SERVING\",\n# \"partitions\": {\n# \"default\": \"SERVING\"\n# }\n# }\n# }\n```\n\n### Delete Collection\n`Client.delete(name: str) -> DashVectorResponse`\n\n| Parameters | Type | Required | Description |\n|------------|-------|----------|---------------------------------------|\n| name | str | Yes | The name of the Collection to delete. |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\nclient.delete('YOUR-COLLECTION-NAME')\n```\n\n### Get a Collection Instance\n`Collection` provides APIs for accessing `Doc` and `Partition`\n\n`Client.get(name: str) -> Collection`\n\n| Parameters | Type | Required | Description |\n|------------|-------|----------|------------------------------------|\n| name | str | Yes | The name of the Collection to get. |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\nassert collection\n```\n\n### Describe Collection Statistics\n`Collection.stats() -> DashVectorResponse`\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\nrsp = collection.stats()\n\nprint(rsp)\n# example output:\n# {\n# \"request_id\": \"14448bcb-c9a3-49a8-9152-0de3990bce59\",\n# \"code\": 0,\n# \"message\": \"Success\",\n# \"output\": {\n# \"total_doc_count\": \"26\",\n# \"index_completeness\": 1.0,\n# \"partitions\": {\n# \"default\": {\n# \"total_doc_count\": \"26\"\n# }\n# }\n# }\n# }\n```\n\n### Insert/Update/Upsert Docs\n```python\nCollection.insert(\n docs: Union[Doc, List[Doc], Tuple, List[Tuple]],\n partition: Optional[str] = None,\n async_req: False\n) -> DashVectorResponse\n```\n\n| Parameters | Type | Required | Description |\n|------------|-------------------------------------------|----------|------------------------------------------------------------------------|\n| docs | Union[Doc, List[Doc], Tuple, List[Tuple]] | Yes | The docs to Insert/Update/Upsert. |\n| partition | Optional[str] | No | Name of the partition to Insert/Update/Upsert.<br/>Default value: `None` |\n| async_req | bool | No | Enable async request or not.<br/>Default value: `False` |\n\nExample:\n```python\nimport dashvector\nimport numpy as np\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\n\n# insert a doc with Tuple\ncollection.insert(('YOUR-DOC-ID1', [0.1, 0.2, 0.3, 0.4]))\ncollection.insert(('YOUR-DOC-ID2', [0.2, 0.3, 0.4, 0.5], {'age': 30, 'name': 'alice', 'anykey': 'anyvalue'}))\n\n# insert a doc with dashvector.Doc\ncollection.insert(\n dashvector.Doc(\n id='YOUR-DOC-ID3', \n vector=[0.3, 0.4, 0.5, 0.6], \n fields={'foo': 'bar'}\n )\n)\n\n# insert in batches\nret = collection.insert(\n [\n ('YOUR-DOC-ID4', [0.2, 0.7, 0.8, 1.3], {'age': 1}),\n ('YOUR-DOC-ID4', [0.3, 0.6, 0.9, 1.2], {'age': 2}),\n ('YOUR-DOC-ID6', [0.4, 0.5, 1.0, 1.1], {'age': 3})\n ]\n)\n\n# insert in batches\nret = collection.insert(\n [\n dashvector.Doc(id=str(i), vector=np.random.rand(4)) for i in range(10)\n ]\n)\n\n# async insert\nret_funture = collection.insert(\n [\n dashvector.Doc(id=str(i+10), vector=np.random.rand(4)) for i in range(10)\n ],\n async_req=True\n)\nret = ret_funture.get()\n```\n\n### Query a Collection\n```python\nCollection.query(\n vector: Optional[Union[List[Union[int, float]], np.ndarray]] = None,\n id: Optional[str] = None,\n topk: int = 10,\n filter: Optional[str] = None,\n include_vector: bool = False,\n partition: Optional[str] = None,\n output_fields: Optional[List[str]] = None,\n async_req: False\n) -> DashVectorResponse\n```\n\n| Parameters | Type | Required | Description |\n|-----------------|------------------------------------------------------|----------|--------------------------------------------------------------------------------------------------------------|\n| vector | Optional[Union[List[Union[int, float]], np.ndarray]] | No | The vector to query |\n| id | Optional[str] | No | The doc id to query.<br/>Setting `id` means searching by vector corresponding to the id |\n| topk | Optional[str] | No | Number of similarity results to return.<br/>Default value: `10` |\n| filter | Optional[str] | No | Expression used to filter results <br/>Default value: None <br/>e.g. `age>20` |\n| include_vector | bool | No | Return vector details or not.<br/>Default value: `False` |\n| partition | Optional[str] | No | Name of the partition to Query.<br/>Default value: `None` |\n| output_fields | Optional[List[str]] | No | List of field names to return.<br/>Default value: `None`, means return all fields<br/>e.g. `['name', 'age']` |\n| async_req | bool | No | Enable async request or not.<br/>Default value: `False` |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\nmatch_docs = collection.query([0.1, 0.2, 0.3, 0.4], topk=100, filter='age>20', include_vector=True, output_fields=['age','name','foo'])\nif match_docs:\n for doc in match_docs:\n print(doc.id)\n print(doc.vector)\n print(doc.fields)\n print(doc.score)\n```\n\n### Delete Docs\n```python\ncollection.delete(\n ids: Union[str, List[str]],\n delete_all: bool = False,\n partition: Optional[str] = None,\n async_req: bool = False\n) -> DashVectorResponse\n```\n\n| Parameters | Type | Required | Description |\n|------------|-----------------------|----------|-----------------------------------------------------------------|\n| ids | Union[str, List[str]] | Yes | The id (or list of ids) for the Doc(s) to Delete |\n| delete_all | bool | No | Delete all vectors from partition.<br/>Default value: `False` |\n| partition | Optional[str] | No | Name of the partition to Delete from.<br/>Default value: `None` |\n| async_req | bool | No | Enable async request or not.<br/>Default value: `False` |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\ncollection.delete(['YOUR-DOC-ID1','YOUR-DOC-ID2'])\n```\n\n### Fetch Docs\n```python\nCollection.fetch(\n ids: Union[str, List[str]],\n partition: Optional[str] = None,\n async_req: bool = False\n) -> DashVectorResponse\n```\n\n| Parameters | Type | Required | Description |\n|------------|-----------------------|----------|----------------------------------------------------------------|\n| ids | Union[str, List[str]] | Yes | The id (or list of ids) for the Doc(s) to Fetch |\n| partition | Optional[str] | No | Name of the partition to Fetch from.<br/>Default value: `None` |\n| async_req | bool | No | Enable async request or not.<br/>Default value: `False` |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\nfetch_docs = collection.fetch(['YOUR-DOC-ID1', 'YOUR-DOC-ID2'])\nif fetch_docs:\n for doc_id in fetch_docs:\n doc = fetch_docs[doc_id]\n print(doc.id)\n print(doc.vector)\n print(doc.fields)\n```\n\n### Create Collection Partition\n`Collection.create_partition(name: str) -> DashVectorResponse`\n\n| Parameters | Type | Required | Description |\n|------------|----------------|----------|-------------------------------------------------------------------------------------------------------|\n| name | str | Yes | The name of the Partition to Create. |\n| timeout | Optional[int] | No | Timeout period (in seconds), -1 means asynchronous creation partition.<br/>Default value: `None` |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\nrsp = collection.create_partition('YOUR-PARTITION-NAME')\nassert rsp\n```\n\n### Delete Collection Partition\n`Collection.delete_partition(name: str) -> DashVectorResponse`\n\n| Parameters | Type | Required | Description |\n|------------|-------|----------|--------------------------------------|\n| name | str | Yes | The name of the Partition to Delete. |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\nrsp = collection.delete_partition('YOUR-PARTITION-NAME')\nassert rsp\n```\n\n### List Collection Partitions\n`Collection.list_partitions() -> DashVectorResponse`\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\npartitions = collection.list_partitions()\n\nassert partitions\nfor pt in partitions:\n print(pt)\n```\n\n### Describe Collection Partition\n`Collection.describe_partition(name: str) -> DashVectorResponse`\n\n| Parameters | Type | Required | Description |\n|------------|-------|----------|----------------------------------------|\n| name | str | Yes | The name of the Partition to Describe. |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\n\nrsp = collection.describe_partition('shoes')\nprint(rsp)\n# example output:\n# {\"request_id\":\"296267a7-68e2-483a-87e6-5992d85a5806\",\"code\":0,\"message\":\"\",\"output\":\"SERVING\"}\n```\n\n### Statistics for Collection Partition\n`Collection.stats_partition(name: str) -> DashVectorResponse`\n\n| Parameters | Type | Required | Description |\n|------------|-------|----------|----------------------------------------------|\n| name | str | Yes | The name of the Partition to get Statistics. |\n\nExample:\n```python\nimport dashvector\n\nclient = dashvector.Client(api_key='YOUR-DASHVECTOR-API-KEY')\ncollection = client.get('YOUR-COLLECTION-NAME')\n\nrsp = collection.stats_partition('shoes')\nprint(rsp)\n# example outptut:\n# {\n# \"code\":0,\n# \"message\":\"\",\n# \"requests_id\":\"330a2bcb-e4a7-4fc6-a711-2fe5f8a24e8c\",\n# \"output\":{\n# \"total_doc_count\":0\n# }\n# }\n```\n\n\n## Class\n### dashvector.Doc\n```python\n@dataclass(frozen=True)\nclass Doc(object):\n id: str\n vector: Union[List[int], List[float], numpy.ndarray]\n fields: Optional[Dict[str, Union[Type[str], Type[int], Type[float], Type[bool]]]] = None \n score: float = 0.0\n```\n\n### dashvector.DashVectorResponse\n\n```python\nclass DashVectorResponse(object):\n code: DashVectorCode\n message: str\n request_id: str\n output: Any\n```\n\n## License\nThis project is licensed under the Apache License (Version 2.0).",
"bugtrack_url": null,
"license": "Apache-2.0",
"summary": "DashVector Client Python Sdk Library",
"version": "1.0.16",
"project_urls": {
"Documentation": "https://help.aliyun.com/document_detail/2510225.html",
"Homepage": "https://github.com/alibaba/proxima"
},
"split_keywords": [
"dashvector",
" vector",
" database",
" cloud"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "ee6a9963913f639ee704bb53cbaeb7bd2a36a1ca55a32ccbbcb428efd16e23fe",
"md5": "88a1be003dd2f8efad4c5d48e02e6352",
"sha256": "734222198d0b6b0317b744055a00c04b4157c3612af1997340b458aaf6f86acf"
},
"downloads": -1,
"filename": "dashvector-1.0.16-py3-none-any.whl",
"has_sig": false,
"md5_digest": "88a1be003dd2f8efad4c5d48e02e6352",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4.0,>=3.7",
"size": 76264,
"upload_time": "2024-06-11T07:51:43",
"upload_time_iso_8601": "2024-06-11T07:51:43.259486Z",
"url": "https://files.pythonhosted.org/packages/ee/6a/9963913f639ee704bb53cbaeb7bd2a36a1ca55a32ccbbcb428efd16e23fe/dashvector-1.0.16-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "00c12d117f6c52dc1c9226891ce4f1f0e35ba3dd32a4f582b42d71ded2de9e66",
"md5": "174300fb9c6e5d5f0cc7694b3ee6f7cc",
"sha256": "e8871c36b79717be82182720365b6b309358f318fdc693b8aacc8adc92c7ba8b"
},
"downloads": -1,
"filename": "dashvector-1.0.16.tar.gz",
"has_sig": false,
"md5_digest": "174300fb9c6e5d5f0cc7694b3ee6f7cc",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4.0,>=3.7",
"size": 46548,
"upload_time": "2024-06-11T07:51:44",
"upload_time_iso_8601": "2024-06-11T07:51:44.692205Z",
"url": "https://files.pythonhosted.org/packages/00/c1/2d117f6c52dc1c9226891ce4f1f0e35ba3dd32a4f582b42d71ded2de9e66/dashvector-1.0.16.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-06-11 07:51:44",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "alibaba",
"github_project": "proxima",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "dashvector"
}
JSON
