https://github.com/Aquila-Network/AquilaPy

Python client library to access Aquila Network Neural Search Engine
https://github.com/Aquila-Network/AquilaPy

aquila-network neural-search personal-search python-client vector-search-engine

Last synced: about 1 month ago
JSON representation

Python client library to access Aquila Network Neural Search Engine

• Host: GitHub
• URL: https://github.com/Aquila-Network/AquilaPy
• Owner: Aquila-Network
• License: mit
• Created: 2020-12-14T13:20:24.000Z (about 4 years ago)
• Default Branch: master
• Last Pushed: 2022-01-13T08:10:52.000Z (almost 3 years ago)
• Last Synced: 2024-10-07T13:37:37.328Z (3 months ago)
• Topics: aquila-network, neural-search, personal-search, python-client, vector-search-engine
• Language: Python
• Homepage: https://aquila.network
• Size: 34.2 KB
• Stars: 3
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        


  

    

  

  


  


    



      

        Aquila Py

      

    

  

  


    

      Python client to access Aquila Network Neural Search Engine

    

  

  




Here is a bird's eye view of where Aquila Client Libraries fit in the entire ecosystem:



  

 




#### install

`pip install aquilapy`

#### Tutorial

```python

from aquilapy import Wallet, DB, Hub

import numpy as np

import time

# Create a wallet instance from private key

wallet = Wallet("private_unencrypted.pem")

host = "http://127.0.0.1"

# Connect to Aquila DB instance

db = DB(host, "5001", wallet)

# Connect to Aquila Hub instance

hub = Hub(host, "5002", wallet)

# Schema definition to be used

schema_def = {

    "description": "this is my database",

    "unique": "r8and0mseEd901",

    "encoder": "strn:msmarco-distilbert-base-tas-b",

    "codelen": 768,

    "metadata": {

        "name": "string",

        "age": "number"

    }

}

# Craete a database with the schema definition provided

db_name = db.create_database(schema_def)

# Craete a database with the schema definition provided

db_name_ = hub.create_database(schema_def)

print(db_name, db_name_)

# Generate encodings

texts = ["Amazon", "Google"]

compression = hub.compress_documents(db_name, texts)

print(compression)

# Prepare documents to be inserted

docs = [{

    "metadata": {

        "name":"name1", 

        "age": 20

    },

    "code": compression[0]

}, {

        "metadata": {

        "name":"name2", 

        "age": 30

    },

    "code": compression[1]

}]

# Insert documents

dids = db.insert_documents(db_name, docs)

print(dids)

# Delete some documents

dids = db.delete_documents(db_name, dids)

print(dids)

# Perform a similarity search operation

matrix = np.random.rand(1, 25).tolist()

time.sleep(5)

docs, dists = db.search_k_documents(db_name, matrix, 10)

print(len(docs[0]), len(dists[0]))

```

created with ❤️ a-mma.indic (a_മ്മ)