API Reference

This section contains the complete API documentation for scikit-agent.

• Blocks
  • Block Classes
  • Model Utilities
• Bellman Periods
  • BellmanPeriod
  • estimate_bellman_foc_residual()
  • estimate_bellman_residual()
  • estimate_discounted_lifetime_reward()
  • estimate_euler_residual()
• Models
  • Consumer Models
  • Benchmark Registry
  • Fisher Two-Period Model
  • Perfect Foresight Models
  • Perfect Foresight Models (Normalized)
  • Resource Extraction
  • Multi-Agent Influence Diagrams (MAIDs)
• Algorithms
  • Value Function Iteration (VFI)
  • Best Response
  • Maliar-Style Algorithms
  • Reinforcement Learning (Stable-Baselines3)
  • Loss Functions
  • Neural Network Components
  • Grid and Computational Tools
• Environments
  • Environment
  • GymEnv
  • discounted_rollout_reward()
• Loss Functions
  • BellmanEquationLoss
  • CustomLoss
  • EstimatedDiscountedLifetimeRewardLoss
  • EulerEquationLoss
  • StaticRewardLoss
  • static_reward()
• Distributions
  • Bernoulli
  • DiscreteDistribution
  • DiscreteDistributionLabeled
  • Distribution
  • IndexDistribution
  • Lognormal
  • MeanOneLogNormal
  • Normal
  • TimeVaryingDiscreteDistribution
  • Uniform
  • combine_indep_dstns()
  • expected()
• Simulation
  • Monte Carlo Simulation
  • Simulation Utility Functions
  • Example Usage
• Parsing
  • ControlToken
  • math_text_to_lambda()
  • skagent_loader()
  • Core Parser Functions
  • Parser Classes
  • Rules
  • Example Usage
• Model Analysis and Visualization
  • Model Analyzer
  • Model Visualizer
  • Influence Diagrams
  • Relevance
• Utils
  • General Utilities
  • Example Usage

Overview

The scikit-agent API is organized into several main modules:

• Blocks - Building blocks – modular model components

• Bellman Periods - Bellman periods and the meta-model notation

• Models - Canonical models from scientific literature

• Algorithms - Solution algorithms and optimization methods

• Environments - Environment adapters for reinforcement-learning algorithms

• Loss Functions - Loss functions for neural network training

• Distributions - Distributions for shocks and initial conditions

• Simulation - Simulation tools and analysis functions

• Parsing - Parsing strings into mathematical expressions

• Model Analysis and Visualization - Model analysis, strategic-relevance graphs, and visualization

• Utils - Utility functions and helpers

Quick Reference

For a quick overview of the most commonly used classes and functions, see the examples in our Examples.