# Base definition of algorithms¶

class orion.algo.base.BaseAlgorithm(space, **kwargs)[source]

Base class describing what an algorithm can do.

Parameters
space`orion.algo.space.Space`

Definition of a problem’s parameter space.

kwargsdict

Tunable elements of a particular algorithm, a dictionary from hyperparameter names to values.

Notes

We are using the No Free Lunch theorem’s [R367dceedad15-1]_[R367dceedad15-3]_ formulation of an `BaseAlgorithm`. We treat it as a part of a procedure which in each iteration suggests a sample of the parameter space of the problem as a candidate solution and observes the results of its evaluation.

Developer Note: Each algorithm’s complete specification, i.e. implementation of its methods and parameters of its own, lies in a separate concrete algorithm class, which must be an immediate subclass of `BaseAlgorithm`. [The reason for this is current implementation of `orion.core.utils.Factory` metaclass which uses BaseAlgorithm.__subclasses__().] Second, one must declare an algorithm’s own parameters (tunable elements which could be set by configuration). This is done by passing them to BaseAlgorithm.__init__() by calling Python’s super with a Space object as a positional argument plus algorithm’s own parameters as keyword arguments. The keys of the keyword arguments passed to BaseAlgorithm.__init__() are interpreted as the algorithm’s parameter names. So for example, a subclass could be as simple as this (regarding the logistics, not an actual algorithm’s implementation):

References

1

D. H. Wolpert and W. G. Macready, “No Free Lunch Theorems for Optimization,” IEEE Transactions on Evolutionary Computation, vol. 1, no. 1, pp. 67–82, Apr. 1997.

2

W. G. Macready and D. H. Wolpert, “What Makes An Optimization Problem Hard?,” Complexity, vol. 1, no. 5, pp. 40–46, 1996.

3

D. H. Wolpert and W. G. Macready, “No Free Lunch Theorems for Search,” Technical Report SFI-TR-95-02-010, Santa Fe Institute, 1995.

Examples

``` 1from orion.algo.base import BaseAlgorithm
2from orion.algo.space import (Integer, Space)
3
4class MySimpleAlgo(BaseAlgorithm):
5
6    def __init__(self, space, multiplier=1, another_param="a string param"):
7        super().__init__(space, multiplier=multiplier, another_param=another_param)
8
9    def suggest(self, num=1):
10        print(self.another_param)
11        return list(map(lambda x: tuple(map(lambda y: self.multiplier * y, x)),
12                        self.space.sample(num)))
13
14    def observe(self, points, results):
15        pass
16
17dim = Integer('named_param', 'norm', 3, 2, shape=(2, 3))
18s = Space()
19s.register(dim)
20
21algo = MySimpleAlgo(s, 2, "I am just sampling!")
22algo.suggest()
```
Attributes
`configuration`

Return tunable elements of this algorithm in a dictionary form appropriate for saving.

`fidelity_index`

Compute the index of the point where fidelity is.

`is_done`

Whether the algorithm is done and will not make further suggestions.

`n_observed`

Number of completed trials observed by the algorithm

`n_suggested`

Number of trials suggested by the algorithm

requires_dist
requires_shape
requires_type
`should_suspend`

Allow algorithm to decide whether a particular running trial is still worth to complete its evaluation, based on information provided by the `judge` method.

`space`

Domain of problem associated with this algorithm’s instance.

`state_dict`

Return a state dict that can be used to reset the state of the algorithm.

Methods

 `format_point`(point) Format point based on space transformations `get_id`(point[, ignore_fidelity]) Compute a unique hash for a point based on params `has_observed`(point) Whether the algorithm has observed a given point objective. `has_suggested`(point) Whether the algorithm has suggested a given point. `judge`(point, measurements) Inform an algorithm about online measurements of a running trial. `observe`(points, results) Observe the results of the evaluation of the points in the process defined in user’s script. `register`(point[, result]) Save the point as one suggested or observed by the algorithm `score`(point) Allow algorithm to evaluate point based on a prediction about this parameter set’s performance. `seed_rng`(seed) Seed the state of the random number generator. `set_state`(state_dict) Reset the state of the algorithm based on the given state_dict `suggest`(num) Suggest a num of new sets of parameters.
property configuration

Return tunable elements of this algorithm in a dictionary form appropriate for saving.

property fidelity_index

Compute the index of the point where fidelity is.

Returns None if there is no fidelity dimension.

format_point(point)[source]

Format point based on space transformations

This will apply the reverse transformation on the point and then transform it again.

Some transformations are lossy and thus the points suggested by the algorithm could be different when returned to `observe`. Using `format_point` makes it possible for the algorithm to see the final version of the point after back and forth transformations. This way it can recognise the point in `observe` and also avoid duplicates that would have gone unnoticed during suggestion.

Parameters
pointtuples of array-likes

Points from a `orion.algo.space.Space`.

get_id(point, ignore_fidelity=False)[source]

Compute a unique hash for a point based on params

Parameters
pointtuples of array-likes

Points from a `orion.algo.space.Space`.

ignore_fidelity: bool, optional

If True, the fidelity dimension is ignored when computing a unique hash for the trial. Defaults to False.

has_observed(point)[source]

Whether the algorithm has observed a given point objective.

This only counts observed completed trials.

Parameters
pointtuples of array-likes

Points from a `orion.algo.space.Space`.

Returns
bool

True if the point’s objective was observed by the algo, False otherwise.

has_suggested(point)[source]

Whether the algorithm has suggested a given point.

Parameters
pointtuples of array-likes

Points from a `orion.algo.space.Space`.

Returns
bool

True if the point was suggested by the algo, False otherwise.

property is_done

Whether the algorithm is done and will not make further suggestions.

Return True, if an algorithm holds that there can be no further improvement. By default, the cardinality of the specified search space will be used to check if all possible sets of parameters has been tried.

judge(point, measurements)[source]

Inform an algorithm about online measurements of a running trial.

Parameters

point – A tuple which specifies the values of the (hyper)parameters used to execute user’s script with.

This method is to be used as a callback in a client-server communication between user’s script and a orion’s worker using a `BaseAlgorithm`. Data returned from this method must be serializable and will be used as a response to the running environment. Default response is None.

Note

Calling algorithm to `judge` a point based on its online measurements will effectively change a state in the algorithm (like a reinforcement learning agent’s hidden state or an automatic early stopping mechanism’s regression), which it may change the value of the property `should_suspend`.

Returns

None or a serializable dictionary containing named data

property n_observed

Number of completed trials observed by the algorithm

property n_suggested

Number of trials suggested by the algorithm

observe(points, results)[source]

Observe the results of the evaluation of the points in the process defined in user’s script.

Parameters
pointslist of tuples of array-likes

Points from a `orion.algo.space.Space`.

resultslist of dicts

Contains the result of an evaluation; partial information about the black-box function at each point in params.

register(point, result=None)[source]

Save the point as one suggested or observed by the algorithm

Parameters
pointarray-likes

Point from a `orion.algo.space.Space`.

resultdict or None, optional

The result of an evaluation; partial information about the black-box function at each point in params. None is suggested and not yet completed.

score(point)[source]

By default, return the same score any parameter (no preference).

Returns

A subjective measure of expected perfomance.

Return type

float

seed_rng(seed)[source]

Seed the state of the random number generator.

Parameters

seed – Integer seed for the random number generator.

Note

This methods does nothing if the algorithm is deterministic.

set_state(state_dict)[source]

Reset the state of the algorithm based on the given state_dict

Parameters

state_dict – Dictionary representing state of an algorithm

property should_suspend

Allow algorithm to decide whether a particular running trial is still worth to complete its evaluation, based on information provided by the `judge` method.

property space

Domain of problem associated with this algorithm’s instance.

property state_dict

Return a state dict that can be used to reset the state of the algorithm.

abstract suggest(num)[source]

Suggest a num of new sets of parameters.

Parameters
num: int

Number of points to suggest. The algorithm may return less than the number of points requested.

Returns
list of points or None

A list of lists representing points suggested by the algorithm. The algorithm may opt out if it cannot make a good suggestion at the moment (it may be waiting for other trials to complete), in which case it will return None.

Notes

New parameters must be compliant with the problem’s domain `orion.algo.space.Space`.