How jbei/kineticlearning works
Kinetic Learning is a Python toolkit and notebook-based workflow for inferring and analyzing dynamical models from time-series measurements. Users prepare or load time-series data, estimate state derivatives, fit learned dynamics, and simulate trajectories; notebooks also support simulated-data creation and domain analyses. The package exposes these stages, while a separate plotting module provides model and learning-curve visualizations. README claims establish notebook capabilities, but the notebook bodies were not sampled, so their internal calls and links to particular data files are not asserted.
Main entry points: __init__.py Data preparation · LearnLimoneneDynamics.ipynb Model analysis · Create Data Set.ipynb Create datasets · data Time-series data
Components 15
Notebook workflows
- Create datasets
notebooks/Create Data Set.ipynb - Model analysis
notebooks/LearnLimoneneDynamics.ipynb
Dynamics modeling
- Data preparation
KineticLearning/__init__.py - Data augmentation
KineticLearning/__init__.py - Derivative estimates
KineticLearning/__init__.py - Model search
KineticLearning/__init__.py - Dynamic model
KineticLearning/__init__.py - Trajectory simulation
KineticLearning/__init__.py - ODE integrator
KineticLearning/__init__.py
Analysis and visualization
- Derivative checks
KineticLearning/__init__.py - Plot utilities
KineticLearning/plot.py - Legacy fit analysis
KineticLearning/helper.py
Other components
- Researcher
- Time-series data
data - TPOT regression
Connections 16
- Researcher → Create datasets: runs
README.md - Researcher → Model analysis: runs
README.md - Create datasets → Time-series data: creates data
- Model analysis → Data preparation: builds models
README.md - Time-series data → Data preparation: loads CSV
KineticLearning/__init__.py - Data preparation → Data augmentation: augments
KineticLearning/__init__.py - Data preparation → Derivative estimates: estimates
KineticLearning/__init__.py - Data preparation → Model search: returns data
KineticLearning/__init__.py - Model search → Dynamic model: fits
KineticLearning/__init__.py - Model search → TPOT regression: searches models
KineticLearning/__init__.py - Dynamic model → Derivative estimates: trains on
KineticLearning/__init__.py - Dynamic model → Trajectory simulation: predicts in
KineticLearning/__init__.py - Trajectory simulation → ODE integrator: integrates with
KineticLearning/__init__.py - Derivative checks → ODE integrator: reconstructs with
KineticLearning/__init__.py - Model analysis → Plot utilities: analyzes with
- Legacy fit analysis → Dynamic model: fits and simulates
KineticLearning/helper.py