mvpa2.clfs.smlr.SMLR¶

class mvpa2.clfs.smlr.SMLR(**kwargs)¶

Sparse Multinomial Logistic Regression Classifier.

This is an implementation of the SMLR algorithm published in Krishnapuram et al., 2005 (2005, IEEE Transactions on Pattern Analysis and Machine Intelligence). Be sure to cite that article if you use this classifier for your work.

Notes

Available conditional attributes:

calling_time+: Time (in seconds) it took to call the node
estimates+: Internal classifier estimates the most recent predictions are based on
predicting_time+: Time (in seconds) which took classifier to predict
predictions+: Most recent set of predictions
raw_results: Computed results before invoking postproc. Stored only if postproc is not None.
trained_dataset: The dataset it has been trained on
trained_nsamples+: Number of samples it has been trained on
trained_targets+: Set of unique targets (or any other space) it has been trained on (if present in the dataset trained on)
training_stats: Confusion matrix of learning performance
training_time+: Time (in seconds) it took to train the learner

(Conditional attributes enabled by default suffixed with +)

Attributes

`auto_train`	Whether the Learner performs automatic trainingwhen called untrained.
`biases`
`descr`	Description of the object if any
`force_train`	Whether the Learner enforces training upon every call.
`pass_attr`	Which attributes of the dataset or self.ca to pass into result dataset upon call
`postproc`	Node to perform post-processing of results
`space`	Processing space name of this node
`trained`	Either classifier was already trained
`weights`

Methods

`__call__`(ds)
`clone`()	Create full copy of the classifier.
`generate`(ds)	Yield processing results.
`get_postproc`()	Returns the post-processing node or None.
`get_sensitivity_analyzer`(\\kwargs)	Returns a sensitivity analyzer for SMLR.
`get_space`()	Query the processing space name of this node.
`is_trained`([dataset])	Either classifier was already trained.
`predict`(obj, data, \args, \\*kwargs)
`repredict`(obj, data, \args, \\*kwargs)
`reset`()
`retrain`(dataset, \\kwargs)	Helper to avoid check if data was changed actually changed
`set_postproc`(node)	Assigns a post-processing node
`set_space`(name)	Set the processing space name of this node.
`summary`()	Providing summary over the classifier
`train`(ds)	The default implementation calls `_pretrain()`, `_train()`, and finally `_posttrain()`.
`untrain`()	Reverts changes in the state of this node caused by previous training

Initialize an SMLR classifier.

Parameters:

lm : float, optional

The penalty term lambda. Larger values will give rise to more sparsification. Constraints: value must be convertible to type ‘float’, and value must be in range [1e-10, inf]. [Default: 0.1]

convergence_tol : float, optional

When the weight change for each cycle drops below this value the regression is considered converged. Smaller values lead to tighter convergence. Constraints: value must be convertible to type ‘float’, and value must be in range [1e-10, 1.0]. [Default: 0.001]

resamp_decay : float, optional

Decay rate in the probability of resampling a zero weight. 1.0 will immediately decrease to the min_resamp from 1.0, 0.0 will never decrease from 1.0. Constraints: value must be convertible to type ‘float’, and value must be in range [0.0, 1.0]. [Default: 0.5]

min_resamp : float, optional

Minimum resampling probability for zeroed weights. Constraints: value must be convertible to type ‘float’, and value must be in range [1e-10, 1.0]. [Default: 0.001]

maxiter : int, optional

Maximum number of iterations before stopping if not converged. Constraints: value must be convertible to type ‘int’, and value must be in range [1, inf]. [Default: 10000]

has_bias : bool, optional

Whether to add a bias term to allow fits to data not through zero. Constraints: value must be convertible to type bool. [Default: True]

fit_all_weights : bool, optional

Whether to fit weights for all classes or to the number of classes minus one. Both should give nearly identical results, but if you set fit_all_weights to True it will take a little longer and yield weights that are fully analyzable for each class. Also, note that the convergence rate may be different, but convergence point is the same. Constraints: value must be convertible to type bool. [Default: True]

implementation : {C, Python}, optional

Use C or Python as the implementation of stepwise_regression. C version brings significant speedup thus is the default one. Constraints: value must be one of (‘C’, ‘Python’). [Default: ‘C’]

ties : str, optional

Resolve ties which could occur. At the moment only obvious ties resulting in identical weights per two classes are detected and resolved randomly by injecting small amount of noise into the estimates of tied categories. Set to False to avoid this behavior. Constraints: value must be a string. [Default: ‘random’]

seed : None or int, optional

Seed to be used to initialize random generator, might be used to replicate the run. Constraints: value must be None, or value must be convertible to type ‘int’. [Default: 866409940]

unsparsify : bool, optional

*EXPERIMENTAL* Whether to unsparsify the weights via regression. Note that it likely leads to worse classifier performance, but more interpretable weights. Constraints: value must be convertible to type bool. [Default: False]

std_to_keep : float, optional

Standard deviation threshold of weights to keep when unsparsifying. Constraints: value must be convertible to type ‘float’. [Default: 2.0]

enable_ca : None or list of str

Names of the conditional attributes which should be enabled in addition to the default ones

disable_ca : None or list of str

Names of the conditional attributes which should be disabled

auto_train : bool

Flag whether the learner will automatically train itself on the input dataset when called untrained.

force_train : bool

Flag whether the learner will enforce training on the input dataset upon every call.

space : str, optional

Name of the ‘processing space’. The actual meaning of this argument heavily depends on the sub-class implementation. In general, this is a trigger that tells the node to compute and store information about the input data that is “interesting” in the context of the corresponding processing in the output dataset.

pass_attr : str, list of str|tuple, optional

Additional attributes to pass on to an output dataset. Attributes can be taken from all three attribute collections of an input dataset (sa, fa, a – see Dataset.get_attr()), or from the collection of conditional attributes (ca) of a node instance. Corresponding collection name prefixes should be used to identify attributes, e.g. ‘ca.null_prob’ for the conditional attribute ‘null_prob’, or ‘fa.stats’ for the feature attribute stats. In addition to a plain attribute identifier it is possible to use a tuple to trigger more complex operations. The first tuple element is the attribute identifier, as described before. The second element is the name of the target attribute collection (sa, fa, or a). The third element is the axis number of a multidimensional array that shall be swapped with the current first axis. The fourth element is a new name that shall be used for an attribute in the output dataset. Example: (‘ca.null_prob’, ‘fa’, 1, ‘pvalues’) will take the conditional attribute ‘null_prob’ and store it as a feature attribute ‘pvalues’, while swapping the first and second axes. Simplified instructions can be given by leaving out consecutive tuple elements starting from the end.

postproc : Node instance, optional

Node to perform post-processing of results. This node is applied in __call__() to perform a final processing step on the to be result dataset. If None, nothing is done.

descr : str

Description of the instance

Attributes

`auto_train`	Whether the Learner performs automatic trainingwhen called untrained.
`biases`
`descr`	Description of the object if any
`force_train`	Whether the Learner enforces training upon every call.
`pass_attr`	Which attributes of the dataset or self.ca to pass into result dataset upon call
`postproc`	Node to perform post-processing of results
`space`	Processing space name of this node
`trained`	Either classifier was already trained
`weights`

Methods

`__call__`(ds)
`clone`()	Create full copy of the classifier.
`generate`(ds)	Yield processing results.
`get_postproc`()	Returns the post-processing node or None.
`get_sensitivity_analyzer`(\\kwargs)	Returns a sensitivity analyzer for SMLR.
`get_space`()	Query the processing space name of this node.
`is_trained`([dataset])	Either classifier was already trained.
`predict`(obj, data, \args, \\*kwargs)
`repredict`(obj, data, \args, \\*kwargs)
`reset`()
`retrain`(dataset, \\kwargs)	Helper to avoid check if data was changed actually changed
`set_postproc`(node)	Assigns a post-processing node
`set_space`(name)	Set the processing space name of this node.
`summary`()	Providing summary over the classifier
`train`(ds)	The default implementation calls `_pretrain()`, `_train()`, and finally `_posttrain()`.
`untrain`()	Reverts changes in the state of this node caused by previous training

biases¶

get_sensitivity_analyzer(**kwargs)¶: Returns a sensitivity analyzer for SMLR.

weights¶