neo-pusher 1.0.1.dev144

A package for Neo4j data ingestion using an AI agent.

neo-pusher

neo-pusher is a Python package designed to facilitate the seamless transfer of data from CSV files to a Neo4j database. Leveraging the power of OpenAI's GPT models and the LangChain framework, neo-pusher automates the process of schema generation, data preprocessing, and data insertion into Neo4j, ensuring data consistency and integrity.

Features

• Automated Schema Generation: Automatically generates a Neo4j schema based on the CSV file's headers.
• Data Preprocessing: Identifies and resolves inconsistencies in the data before pushing it to Neo4j.
• Error Handling and Debugging: The agent reruns and debugs code automatically if any errors are encountered during the process.
• Multiple Dataset Support: Capable of handling multiple datasets simultaneously, ensuring that all columns are properly represented in the database.

Installation

To install neo-pusher, use pip:

pip install neo-pusher

Usage

Here's an example of how to use neo-pusher to push data from a CSV file to a Neo4j database:

from neo_pusher.agent import NeoAgent

# Initialize the NeoAgent with your OpenAI API key
agent = NeoAgent(apikey="your_openai_api_key",lang_chain_api_key="your_lang_chain_api_key")

# Define the parameters for your Neo4j database and CSV file
path = "path/to/your/csvfile.csv"
username = "neo4j_username"
password = "neo4j_password"
url = "bolt://localhost:7687"
data = "head of your csv data"

# Run the agent to push data to Neo4j
response = agent.run(path, username, password, url, data)
print(response)

Parameters

• apikey (str): Your OpenAI API key.
• langchain_api_key (str): Your Langchain API key.
• model (str): The OpenAI model to use. Defaults to "gpt-4o".
• path (str): The path to the CSV file.
• username (str): The username for the Neo4j database.
• password (str): The password for the Neo4j database.
• url (str): The URL of the Neo4j database.
• data (str): The head of the CSV file. Defaults to None.

Return Value

• The run method returns a response from the LLM, including the results of the schema generation, data preprocessing, and data insertion into Neo4j.

Example CSV Data

The following is an example of the CSV file headers that neo-pusher can process:

Dataset 1:

order_details_id	order_id	pizza_id	quantity
1	1	hawaiian_m	1
2	2	classic_dlx_m	1

Dataset 2:

pizza_id	pizza_type_id	size	price
bbq_ckn_s	bbq_ckn	S	12.75
bbq_ckn_m	bbq_ckn	M	16.75

Notes

• The agent uses the neo4j Python package to connect to and push data into the Neo4j database.
• Before pushing data, the agent checks for any inconsistencies and cleans the data accordingly.