mvpa2.testing.clfs.SMLR¶

class
mvpa2.testing.clfs.
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+
: Noneestimates+
: Internal classifier estimates the most recent predictions are based onpredicting_time+
: Time (in seconds) which took classifier to predictpredictions+
: Most recent set of predictionsraw_results
: Nonetrained_dataset
: Nonetrained_nsamples+
: Nonetrained_targets+
: Nonetraining_stats
: Confusion matrix of learning performancetraining_time+
: None
(Conditional attributes enabled by default suffixed with
+
)Methods
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 [1e10, 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 [1e10, 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 [1e10, 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: 1984161704]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 subclass 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 strtuple, 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 postprocessing 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
Methods

biases
¶

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

weights
¶