CohortSpec

models.CohortSpec(
    transition,
    state_cost,
    state_effect,
    transition_cost=None,
    transition_effect=None,
)

One intervention’s matrices for a single parameter set.

Returned by the engine’s transitions_and_rewards function. Arrays are plain numpy arrays over the engine’s state order.

Parameters

Name	Type	Description	Default
transition	NDArray[np.float64]	Transition-probability matrix, shape (n_states, n_states) for a time-independent model or (n_cycles, n_states, n_states) when transitions vary by cycle. Each row sums to 1.	required
state_cost	NDArray[np.float64]	Per-cycle cost of occupying each state, shape (n_states,), or (n_cycles + 1, n_states) when it varies.	required
state_effect	NDArray[np.float64]	Per-cycle effect (QALYs) of each state, same shape rule as state_cost.	required
transition_cost	NDArray[np.float64] | None	Optional extra cost attached to a transition, added on the flow between states, shape (n_states, n_states) or (n_cycles, n_states, n_states). Entry [i, j] is the cost of moving from state i to state j.	None
transition_effect	NDArray[np.float64] | None	Optional effect attached to a transition, same shape rule as transition_cost.	None