14 hours (usually 2 days including breaks)
- Python programming experience
Mesa is a Python framework for creating agent-based models (ABM). Mesa aims to provide an alternative to NetLogo, Repast, and MASON. Agent based modeling allows investigators in the fields of biology, social sciences, network, business, etc. to simulate the actions and interactions of autonomous agents in order to evaluate their effects on their environment.
This instructor-led, live training (online or onsite) is aimed at investigators who wish to use Mesa to create Agent Based Models in a Python environment.
By the end of this training, participants will be able to:
- Install and configure the development environment needed to start modeling in Python.
- Quickly create an agent-based model using Mesa's built-in core components.
- Expand the complexity of the model.
- Visualize agent activity in real-time inside a browser.
- Analyze the results of the model interactively using Python data analysis tools.
- Integrate the model with other Python systems such as machine learning applications.
Format of the Course
- Interactive lecture and discussion.
- Lots of exercises and practice.
- Hands-on implementation in a live-lab environment.
Course Customization Options
- To request a customized training for this course, please contact us to arrange.
Overview of Agent Based Modeling
Case Study: Using Agents to Simulate Financial Transactions
Overview of Agent Based Modeling Frameworks for Java, C++, Python, etc.
Overview of Mesa's Core Features
Setting up the Environment
Choosing between a Text Editor or IDE and Jupyter Notebook
Creating a Simple Model
Case Study: Using Agents to Simulate a Pandemic
Choosing a Model Based on the Use Case (Boltzmann Wealth, Schelling Segregation Model, SIR, etc.)
Working with the Mesa's Model and Agent Classes
Defining the Variables
Setting Model Level Parameters
Scheduling the Actions of an Agent
Running the Model
Adding Agents to the Model
Adding Space to the Model
Collecting Data Using the Data Collector
Running the Model Multiple Using the Mesa Batch Runner
Visualizing the Simulation Interactively
Visualizing Agent Activity in a Grid
Adding a Chart to the Visualization
Integrating the Model with a Machine Learning Application.
Summary and Conclusion