Saving and Loading Models

The epimodels.io module serializes model specifications — family, class name, state variables, parameter symbols, the last-run parameter values and (optionally) simulation traces — to JSON or YAML files. Reconstruction goes through the model registry, so saved models round-trip across sessions and machines.

Saving and loading

from epimodels.continuous import SIR
from epimodels.io import save_model, load_model

model = SIR()
model([1000, 1, 0], [0, 50], 1001, {"beta": 2, "gamma": 0.1})

save_model(model, "sir_run.json", include_traces=True)
clone = load_model("sir_run.json")

clone.param_values["beta"]   # 2
clone.traces["I"]            # restored as numpy arrays

YAML requires the optional yaml extra (pip install epimodels[yaml]):

save_model(model, "sir_run.yaml")

Serialization also works for discrete, CTMC and network models:

from epimodels.network import NetworkSIR
import networkx as nx

net = NetworkSIR(nx.barabasi_albert_graph(500, 3, seed=0))
save_model(net, "network.json")   # structure spec (graph not serialized)

Dict-based API

For programmatic use (e.g. storing specs in a database), the same functionality is available as dictionaries:

from epimodels.io import model_to_dict, model_from_dict

spec = model_to_dict(model, include_traces=True)
clone = model_from_dict(spec)

Note:

The graph object of network models is not serialized — only the model class and parameters are. Rebuild the graph (or pass it again) when re-running network simulations.