plada 0.0.8

PlaDa: Platform for Data market

plada

PlaDa (Platform for Data market)

PlaDa is a simulation platform designed to model and analyze data-driven marketplaces. It provides tools for setting up market models, configuring simulation parameters, running simulations, and analyzing results.

Install

PlaDa is available on PyPI and can be installed using pip.

Using pip

$ pip install plada
$ python
>> import plada

Using pip in a Jupyter Notebook

When installing within a Jupyter Notebook, use the !pip command:

!pip install plada
import plada

For more detailed examples, refer to the test.ipynb notebook.

Usage

Step1：Set up the Market Model.

Note: The model must be a Graph object and must contain "variables".

model = nx.read_graphml("test.graphml")

Step2：Configure Settings

Define the configuration for the simulation, market, and agents.

config = {
    "Simulation":{
        "num_iterations": 10,
        "num_steps": 10,
        "isPrice": True,
    },
    "Market":{
        "model": model,
    },
    "Agent": {
        "num_buyers": 10,
        "strategy_weights": {
            "random": 0.0,
            "related": 0.0,
            "ranking": 1.0,
        },
        "new_buyer_probability": 0.8,
    }
}

Step3：Run the Simulation

Initialize the Saver and Runner classes with the configuration settings and logger. Then, execute the main simulation process.

saver = Saver()
runner = Runner(settings=config, logger=saver)

runner.main()

Explanation

buyer.py

• Configuration of Buyer Agent
• Manages the state and strategies of buyers, updates budget, and saves purchased data.

market.py

• Market Class: Manages the overall market.
• Data Class: Manages the data.
• Variable Class: Manages the variables.
• Primarily responsible for updating prices and related market dynamics.

simulator.py

• Setup of the Purchase Simulation Workflow
• Organizes the information to be logged (requires improvement)

runner.py

• Execution of the Simulation

logger.py

• Saving the Results of the Simulation

analyzer.py

• Basic Analysis of the Results
• Visualizes the distribution of purchase counts.