rsr.rsr.get_comp_cond_sys_prob_multi

rsr.rsr.get_comp_cond_sys_prob_multi(refs_dict_upper, refs_dict_lower, probs, comps_st_cond, row_names, s_fun=None, n_sample=1000000, n_batch=1000000)[source]

Estimate P(system state = s | given component states) for multi-state systems by Monte Carlo.

Parameters:
  • refs_dict_upper (Dict[int, Tensor]) – dict of system upper reference state tensors {state: Tensor(n_var, n_state)}.

  • refs_dict_lower (Dict[int, Tensor]) – dict of system lower reference state tensors {state: Tensor(n_var, n_state)}.

  • probs (Tensor) – (n_var, n_state) categorical probability tensor.

  • comps_st_cond (Dict[str, int]) – dict of known component states {name: state_index}.

  • row_names (Sequence[str]) – list of variable (component) names matching probs rows.

  • s_fun (callable) – function(comps_dict) -> tuple(_, sys_state, _).

  • n_sample (int) – number of samples total and per batch.

  • n_batch (int) – number of samples total and per batch.

Returns:

probability}, summing to 1.0.

Return type:

Dictionary {state