aepsych.strategy¶
aepsych.strategy module¶
- class aepsych.strategy.Strategy(generator, lb, ub, stimuli_per_trial, outcome_types, dim=None, min_total_tells=0, min_asks=0, model=None, refit_every=1, min_total_outcome_occurrences=1, max_asks=None, keep_most_recent=None, min_post_range=None, name='', run_indefinitely=False)[source]¶
Bases:
object
Object that combines models and generators to generate points to sample.
Initialize the strategy object.
- Parameters:
generator (AEPsychGenerator) – The generator object that determines how points are sampled.
lb (Union[numpy.ndarray, torch.Tensor]) – Lower bounds of the parameters.
ub (Union[numpy.ndarray, torch.Tensor]) – Upper bounds of the parameters.
dim (int, optional) – The number of dimensions in the parameter space. If None, it is inferred from the size of lb and ub.
min_total_tells (int) – The minimum number of total observations needed to complete this strategy.
min_asks (int) – The minimum number of points that should be generated from this strategy.
model (ModelProtocol, optional) – The AEPsych model of the data.
refit_every (int) – How often to refit the model from scratch.
min_total_outcome_occurrences (int) – The minimum number of total observations needed for each outcome before the strategy will finish. Defaults to 1 (i.e., for binary outcomes, there must be at least one “yes” trial and one “no” trial).
max_asks (int, optional) – The maximum number of trials to generate using this strategy. If None, there is no upper bound (default).
keep_most_recent (int, optional) – Experimental. The number of most recent data points that the model will be fitted on. This may be useful for discarding noisy data from trials early in the experiment that are not as informative as data collected from later trials. When None, the model is fitted on all data.
min_post_range (float, optional) – Experimental. The required difference between the posterior’s minimum and maximum value in probablity space before the strategy will finish. Ignored if None (default).
name (str) – The name of the strategy. Defaults to the empty string.
run_indefinitely (bool) – If true, the strategy will run indefinitely until finish() is explicitly called. Other stopping criteria will be ignored. Defaults to False.
stimuli_per_trial (int) –
outcome_types (Sequence[Type[str]]) –
- normalize_inputs(x, y)[source]¶
converts inputs into normalized format for this strategy
- Parameters:
x (np.ndarray) – training inputs
y (np.ndarray) – training outputs
- Returns:
training inputs, normalized y (np.ndarray): training outputs, normalized n (int): number of observations
- Return type:
x (np.ndarray)
- property finished¶
- property can_fit¶
- property n_trials¶
- class aepsych.strategy.SequentialStrategy(strat_list)[source]¶
Bases:
object
Runs a sequence of strategies defined by its config
All getter methods defer to the current strat
- Parameters:
strat_list (list[Strategy]) – TODO make this nicely typed / doc’d
- property finished¶
- class aepsych.strategy.AEPsychStrategy(ax_client)[source]¶
Bases:
ConfigurableMixin
- Parameters:
ax_client (AxClient) –
- is_finished = False¶
- classmethod get_config_options(config, name=None)[source]¶
- Parameters:
config (Config) –
name (Optional[str]) –
- Return type:
Dict
- property finished: bool¶
- property experiment¶
- property strat¶
- property can_fit¶
- plot_contours(density=50, slice_values=None)[source]¶
Plot predictions for a 2-d slice of the parameter space.
- Parameters:
density (int) – Number of points along each parameter to evaluate predictions.
slice_values (Optional[Dict[str, Any]]) – A dictionary {name: val} for the fixed values of the other parameters. If not provided, then the mean of numeric parameters or the mode of choice parameters will be used.
- plot_slice(param_name, density=50, slice_values=None)[source]¶
Plot predictions for a 1-d slice of the parameter space.
- Parameters:
param_name (str) – Name of parameter that will be sliced
density (int) – Number of points along slice to evaluate predictions.
slice_values (Optional[Dict[str, Any]]) – A dictionary {name: val} for the fixed values of the other parameters. If not provided, then the mean of numeric parameters or the mode of choice parameters will be used.