bsfit.nodes.models.utils.get_trial_prediction

bsfit.nodes.models.utils.get_trial_prediction(output: dict, params: dict, n_repeats: int)

get model-generated stochastic choices

Parameters

• output (dict) –

  'PestimateGivenModel': estimate probabilities
  'map': max-a-posteriori percepts
  'conditions': task conditions
  

• params (dict) –

  parameters:

  params: {
      "task": {
          "fixed_params": the task fixed parameters
          },
      "model": {
          "fixed_params": the model fixed parameters
          "init_params": the model initial parameters
      }
  

• n_repeats (int) – repeat count per condition, when granularity=”trial”

Returns

returns “dataset” to output

Return type

(dict)

Note

-for now “stim_mean” must be integers