Response¶

dimod samplers respond with a consistent Response object that is Iterable (over samples, from lowest energy to highest) and Sized (number of samples).

Examples

This example shows the response of the dimod ExactSolver sampler.

>>> import dimod
>>> response = dimod.ExactSolver().sample_ising({'a': -0.5}, {})
>>> len(response)
2
>>> for sample in response:
...     print(sample)
{'a': 1}
{'a': -1}

Class¶

class Response(**kwargs)[source]¶

A container for samples and any other data returned by dimod samplers.

Parameters:

samples_matrix (numpy.matrix) – Samples as a NumPy matrix where each row is a sample.
data_vectors (dict[field, numpy.array/list]) – Additional per-sample data as a dict of vectors. Each vector is of the same length as samples_matrix. The key ‘energy’ and its vector are required.
vartype (Vartype) – Vartype of the samples.
info (dict, optional, default=None) – Information about the response as a whole formatted as a dict.
variable_labels (list, optional, default=None) – Variable labels mappped by index to columns of the samples matrix.

vartype¶: Vartype – Vartype of the samples.

info¶: dict – Information about the response as a whole formatted as a dict.

variable_labels¶: list/None – Variable labels. Each column in the samples matrix is the values returned for one variable. If None, column indices are the labels.

label_to_idx¶: dict – Map of variable labels to columns in samples matrix.

Examples

This example shows some attributes of the response for the sampler of dimod package’s random_sampler.py reference example.

>>> from dimod.reference.samplers.random_sampler import RandomSampler
>>> sampler = RandomSampler()
>>> bqm = dimod.BinaryQuadraticModel({0: 0.0, 1: 1.0}, {(0, 1): 0.5}, -0.5, dimod.SPIN)
>>> response = sampler.sample(bqm)
>>> response.vartype  
<Vartype.SPIN: frozenset([1, -1])>  
>>> response.variable_labels
[0, 1]

Properties¶

`Response.samples_matrix`	`numpy.matrix` – Samples as a NumPy matrix of data type int8.
`Response.data_vectors`	dict[field, `numpy.array`/list] – Per-sample data as a dict, where keys are the data labels and values are each a vector of the same length as sample_matrix.

Methods¶

Viewing a Response¶

`Response.samples`([n, sorted_by])	Iterate over the samples in the response.
`Response.data`([fields, sorted_by, name])	Iterate over the data in the response.

Constructing or Updating a Response¶

`Response.from_dicts`(samples, data_vectors[, …])	Build a response from an iterable of dicts.
`Response.from_futures`(futures, vartype, …)	Build a response from `Future`-like objects.
`Response.from_matrix`(samples, data_vectors)	Build a response from a NumPy array-like object.
`Response.from_pandas`(samples_df, data_vectors)	Build a response from a pandas DataFrame.
`Response.update`(*other_responses)	Add values of other responses to the response.

Transformations¶

`Response.change_vartype`(**kwargs)	Create a new response with the given vartype.
`Response.relabel_variables`(mapping[, inplace])	Relabel a response’s variables as per a given mapping.

Copy¶

Response.copy() Creates a shallow copy of a response.

Done¶

Response.done() True if all loaded futures are done or if there are no futures.