mvpa2.clfs.sg.SVM

class mvpa2.clfs.sg.SVM(**kwargs)

Support Vector Machine Classifier(s) based on Shogun

This is a simple base interface

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.
descr	Description of the object if any
force_train	Whether the Learner enforces training upon every call.
kernel_params
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
svm	Access to the SVM model.
traindataset	Dataset which was used for training
trained	Either classifier was already trained

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 an appropriate SensitivityAnalyzer.
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

Interface class to Shogun’s classifiers and regressions.

Default implementation is ‘libsvm’.

SVM/SVR definition is dependent on specifying kernel, implementation type, and parameters for each of them which vary depending on the choices made.

Desired implementation is specified in svm_impl argument. Here is the list if implementations known to this class, along with specific to them parameters (described below among the rest of parameters), and what tasks it is capable to deal with (e.g. regression, binary and/or multiclass classification):

libsvr

: LIBSVM’s epsilon-SVR

Parameters: C, tube_epsilon Capabilities: regression

gnpp

: Generalized Nearest Point Problem SVM

Parameters: C Capabilities: binary

libsvm

: LIBSVM’s C-SVM (L2 soft-margin SVM)

Parameters: C Capabilities: binary, multiclass

gmnp

: Generalized Nearest Point Problem SVM

Parameters: C Capabilities: binary, multiclass

gpbt

: Gradient Projection Decomposition Technique for large-scale SVM problems

Parameters: C Capabilities: binary

Kernel choice is specified as a kernel instance with kwargument kernel. Some kernels (e.g. Linear) might allow computation of per feature sensitivity.

Parameters:

tube_epsilon

Epsilon in epsilon-insensitive loss function of epsilon-SVM regression (SVR). [Default: 0.01]

C

Trade-off parameter between width of the margin and number of support vectors. Higher C – more rigid margin SVM. In linear kernel, negative values provide automatic scaling of their value according to the norm of the data. [Default: -1.0]

epsilon

Tolerance of termination criteria. (For nu-SVM default is 0.001). [Default: 5e-05]

kernel

Kernel object. [Default: None]

num_threads

Number of threads to utilize. [Default: 1]

retrainable : bool, optional

Either to enable retraining for ‘retrainable’ classifier. Constraints: value must be convertible to type bool. [Default: False]

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.
descr	Description of the object if any
force_train	Whether the Learner enforces training upon every call.
kernel_params
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
svm	Access to the SVM model.
traindataset	Dataset which was used for training
trained	Either classifier was already trained

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 an appropriate SensitivityAnalyzer.
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

svm

Access to the SVM model.

traindataset

Dataset which was used for training

TODO – might better become conditional attribute I guess