Algorithms
The Splunk Machine Learning Toolkit supports the algorithms listed here. In addition to the examples included in the Splunk Machine Learning Toolkit, you can find more examples of these algorithms on the scikit-learn website.
Looking for information on using the score
command? Please navigate to the score command documentation for more information.
ML-SPL Quick Reference Guide
Download the Machine Learning Toolkit Quick Reference Guide for a handy cheat sheet of ML-SPL commands and machine learning algorithms used in the Splunk Machine Learning Toolkit. This document is also available in Japanese.
Download the ML-SPL Performance App for Machine Learning Toolkit to use these performance results for guidance and benchmarking purposes in your own environment.
Algorithms using the fit
and apply
commands
The following algorithms use the fit
and apply
commands within the Splunk Machine Learning Toolkit. For information on the steps taken by these commands, please review the Understanding the fit and apply commands document.
Extend the algorithms you can use for your models
The algorithms listed here and in the ML-SPL Quick Reference Guide are available natively in the Splunk Machine Learning Toolkit. You can also base your algorithm on over 300 open source Python algorithms from scikit-learn, pandas, statsmodel, numpy and scipy libraries available through the Python for Scientific Computing add-on in Splunkbase.
For information on how to import an algorithm from the Python for Scientific Computing add-on into the Splunk Machine Learning Toolkit, see the ML-SPL API Guide.
Anomaly Detectors
Anomaly detection algorithms detect anomalies and outliers in numerical or categorical fields.
LocalOutlierFactor
The LocalOutlierFactor algorithm uses the scikit-learn Local Outlier Factor (LOF) (an unsupervised outlier detection method) to measure the local deviation of density of a given sample with respect to its neighbors. The anomaly score depends on how isolated the object is with respect to its neighbors.
Syntax
fit LocalOutlierFactor <fields> [n_neighbors=<int>] [leaf_size=<int>] [p=<int>] [contamination=<float>] [metric=<str>] [algorithm=<str>]
- The
n_neighbors
parameter default is 20 - The
leaf_size
parameter default is 30 - The
p
parameter is limited top >=1
- The
contamination
parameter must be within the range of 0.0 (not included) to 0.5 (included) - The
contamination
parameter default is 0.1 - Valid algorithms:
brute
,kd_tree
,ball_tree
andauto
- The
algorithm
default is auto - The
metric
default is minkowski - Valid metrics for kd_tree:
cityblock
,euclidean
,l1
,l2
,manhattan
,chebyshev
,minkowski
- Valid metrics for ball_tree:
cityblock
,euclidean
,l1
,l2
,manhattan
,chebyshev
,minkowski
,braycurtis
,canberra
,dice
,hamming
,jaccard
,kulsinski
,matching
,rogerstanimoto
,russellrao
,sokalmichener
,sokalsneath
- Valid metrics for brute:
cityblock
,euclidean
,l1
,l2
,manhattan
,chebyshev
,minkowski
,braycurtis
,canberra
,dice
hamming
,jaccard
,kulsinski
,matching
,rogerstanimoto
,russellrao
,sokalmichener
,sokalsneath
,cosine
,correlation
,sqeuclidean
,yule
- Output: A list of labels titled
is_outlier
, assigned1
for outliers and-1
for inliers
Example
| inputlookup iris.csv | fit LocalOutlierFactor petal_length petal_width n_neighbors=10 algorithm=kd_tree metric=minkowski p=1 contamination=0.14 leaf_size=10
For descriptions of the n_neighbors
, leaf_size
and other parameters, see the sci-kit learn documentation: http://scikit-learn.org/stable/modules/generated/sklearn.neighbors.LocalOutlierFactor.html
You cannot save LocalOutlierFactor models using the into
keyword. This algorithm does not support saving models. It does not include the predict
method.
Using the LocalOutlierFactor algorithm requires running version 1.3 of Python for Scientific Computing.
OneClassSVM
The OneClassSVM algorithm uses the scikit-learn OneClassSVM (an unsupervised outlier detection method) to fit a model from the set of features (i.e. fields) for detecting anomalies/outliers, where features are expected to contain numerical values.
Syntax
fit OneClassSVM <fields> [into <model name>] [kernel=<str>] [nu=<float>] [coef0=<float>] [gamma=<float>] [tol=<float>] [degree=<int>] [shrinking=<true|false>]
- The
kernel
parameter specifies the kernel type ("linear", "rbf", "poly", "sigmoid") for using in the algorithm, where the default value is kernel isrbf
. * You can specify the upper bound on the fraction of training error as well as the lower bound of the fraction of support vectors using thenu
parameter, where the default value is 0.5. - The
degree
parameter is ignored by all kernels except the polynomial kernel, where the default value is 3.gamma
is the kernel co-efficient that specifies how much influence a single data instance has, where the default value is 1/numberOfFeatures. The independent term ofcoef0
in the kernel function is only significant if you have polynomial or sigmoid function. The termtol
is the tolerance for stopping criteria. - The
shrinking
parameter determines whether to use the shrinking heuristic. For details, see http://scikit-learn.org/stable/modules/svm.html#kernel-functions.
Example
... | fit OneClassSVM * kernel="poly" nu=0.5 coef0=0.5 gamma=0.5 tol=1 degree=3 shrinking=f into TESTMODEL_OneClassSVM
You can save OneClassSVM models using the into
keyword and apply the saved model later to new data using the apply
command. See the example below.
Example
... | apply TESTMODEL_OneClassSVM
After running the fit
or apply
command, a new field named isNormal
is generated. This field defines whether a particular record (row) is normal (isNormal=1
) or anomalous (isNormal=-1
).
You cannot inspect the model learned by OneClassSVM with the summary
command.
Classifiers
Classifier algorithms predict the value of a categorical field.
The kfold
cross-validation command is available for all Classifier algorithms. Learn more here.
BernoulliNB
The BernoulliNB algorithm uses the scikit-learn BernoulliNB estimator (an implementation of the Naive Bayes classification algorithm) to fit a model to predict the value of categorical fields where explanatory variables are assumed to be binary-valued. This algorithm supports incremental fit.
Syntax
fit BernoulliNB <field_to_predict> from <explanatory_fields> [into <model name>] [alpha=<float>] [binarize=<float>] [fit_prior=<true|false>] [partial_fit=<true|false>]
- The
alpha
parameter controls Laplace/ Lidstone smoothing. The default value is 1.0. - The
binarize
parameter is a threshold that can be used for converting numeric field values to the binary values expected by BernoulliNB. The default value is 0. For instance, ifbinarize=0
is specified, the default, values > 0 are assumed to be 1, and values <= 0 are assumed to be 0. - The
fit_prior
Boolean parameter specifies whether to learn class prior probabilities. The default value istrue
. Iffit_prior=f
is specified, classes are assumed to have uniform popularity. - The
partial_fit
parameter controls whether an existing model should be incrementally updated or not. The default value isfalse
, meaning it will not be incrementally updated. Choosingpartial_fit=true
allows you to update an existing model using only new data without having to retrain it on the full training data set.- Using
partial_fit=true
on an existing model ignores the newly supplied parameters. The parameters supplied at model creation are used instead. Ifpartial_fit=false
orpartial_fit
is not specified (default is false), the model specified is created and replaces the pre-trained model if one exists.
- Using
Example 1
... | fit BernoulliNB type from * into TESTMODEL_BernoulliNB alpha=0.5 binarize=0 fit_prior=f
Example 2
In the following example, if My_Incremental_Model
does not exist, the command saves the model data under the model name My_Incremental_Model
. If My_Incremental_Model
exists and was trained using BernoulliNB, the command updates the existing model with the new input. If My_Incremental_Model
exists but was not trained by BernoulliNB, an error message is thrown.
| inputlookup iris.csv | fit BernoulliNB species from * partial_fit=true into My_Incremental_Model
You can save BernoulliNB models using the into
keyword and apply the saved model later to new data using the apply
command. See the following example.
Example
|... | apply TESTMODEL_BernoulliNB
You cannot inspect the model learned by BernoulliNB with the summary
command.
DecisionTreeClassifier
The DecisionTreeClassifier algorithm uses the scikit-learn DecisionTreeClassifier estimator to fit a model to predict the value of categorical fields.
Syntax
fit DecisionTreeClassifier <field_to_predict> from <explanatory_fields> [into <model_name>] [max_depth=<int>] [max_features=<str>] [min_samples_split=<int>] [max_leaf_nodes=<int>] [criterion=<gini|entropy>] [splitter=<best|random>] [random_state=<int>]
Example
... | fit DecisionTreeClassifier SLA_violation from * into sla_model | ...
For descriptions of the max_depth
, max_features
, min_samples_split
, max_leaf_nodes
, criterion
, random_state
, and splitter
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html.
You can save DecisionTreeClassifier models by using the into
keyword and apply it to new data later by using the apply
command. See example below.
Example
... | apply model_DTC
You can inspect the decision tree learned by DecisionTreeClassifier with the summary
command. See example below.
Example
| summary model_DTC
You can also get a JSON representation of the tree by giving json=t
as an argument to the summary
command. See example below.
| summary model_DTC json=t
To specify the maximum depth of the tree to summarize, use the limit
argument. The default value for the limit
argument is 5. See example below.
| summary model_DTC limit=10
GaussianNB
The GaussianNB algorithm uses the scikit-learn GaussianNB estimator (an implementation of Gaussian Naive Bayes classification algorithm) to fit a model to predict the value of categorical fields, where the likelihood of explanatory variables is assumed to be Gaussian. This algorithm supports incremental fit.
Syntax
fit GaussianNB <field_to_predict> from <explanatory_fields> [into <model name>] [partial_fit=<true|false>]
Example
... | fit GaussianNB species from * into TESTMODEL_GaussianNB
The partial_fit
parameter controls whether an existing model should be incrementally updated or not (default is false
). This allows you to update an existing model using only new data without having to retrain it on the full training data set.
Example
The following example uses the partial_fit
command. In this example, if My_Incremental_Model
does not exist, the command saves the model data under the model name My_Incremental_Model
. If My_Incremental_Model
exists and was trained using GaussianNB, the command updates the existing model with the new input. If My_Incremental_Model
exists but was not trained by GaussianNB, an error message is thrown.
| inputlookup iris.csv | fit GaussianNB species from * partial_fit=true into My_Incremental_Model
If partial_fit=false
or partial_fit
is not specified (default is false
), the model specified is created and replaces the pre-trained model if one exists.
You can save GaussianNB models using the into
keyword and apply the saved model later to new data using the apply
command. See the following example:
Example
... | apply TESTMODEL_GaussianNB
You cannot inspect the model learned by GaussianNB with the summary
command.
GradientBoostingClassifier
This algorithm uses the GradientBoostingClassifier from scikit-learn to build a classification model by fitting regression trees on the negative gradient of a deviance loss function. For documentation on the parameters, see GradientBoostingClassifier from scikit-learn http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingClassifier.html.
Syntax
fit GradientBoostingClassifier <field_to_predict> from <explanatory_fields>[into <model name>] [loss=<deviance | exponential>] [max_features=<str>] [learning_rate =<float>] [min_weight_fraction_leaf=<float>] [n_estimators=<int>] [max_depth=<int>] [min_samples_split =<int>] [min_samples_leaf=<int>] [max_leaf_nodes=<int>] [random_state=<int>]
Example
The following example uses the GradientBoostingClassifier algorithm to fit a model and save a model as TESTMODEL_GradientBoostingClassifier
.
... | fit GradientBoostingClassifier target from * into TESTMODEL_GradientBoostingClassifier
Use the apply method to apply the trained model to the new data.
... |apply TESTMODEL_GradientBoostingClassifier
To inspect the features learned by GradientBoostingClassifier use the summary command.
| summary TESTMODEL_GradientBoostingClassifier
LogisticRegression
The LogisticRegression algorithm uses the scikit-learn LogisticRegression estimator to fit a model to predict the value of categorical fields.
Syntax
fit LogisticRegression <field_to_predict> from <explanatory_fields> [into <model name>] [fit_intercept=<true|false>] [probabilities=<true|false>]
Example
... | fit LogisticRegression SLA_violation from IO_wait_time into sla_model | ...
The fit_intercept
parameter specifies whether the model includes an implicit intercept term (the default value is true
).
The probabilities
parameter specifies whether probabilities for each possible field value should be returned alongside the predicted value (the default value is false
).
You can save LogisticRegression models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply sla_model
You can inspect the coefficients learned by LogisticRegression with the summary
command (for example, | summary sla_model
).
MLPClassifier
The MPLClassifier algorithm uses the scikit-learn Multi-layer Perceptron estimator for classification. MLPClassifier uses a feedforward artificial neural network model that trains using backpropagation. This algorithm supports incremental fit.
Syntax
fit MLPClassifier <field_to_predict> from <explanatory_fields> [into <model name>] [batch_size=<int>] [max_iter=<int>] [random_state=<int>] [hidden_layer_sizes=<int>-<int>-<int>] [activation=<str>] [solver=<str>] [learning_rate=<str>] [tol=<float>} {momentum=<float>]
Example
... | inputlookup diabetes.csv | fit MLPClassifier response from * into MLP_example_model hidden_layer_sizes='100-100-80' |...
Example
The following example uses the partial_fit
command:
| inputlookup iris.csv | fit MLPClassifier species from * partial_fit=true into My_Example_Model
The partial_fit
parameter controls whether an existing model should be incrementally updated on not (default is false
). This allows you to update nan existing model using only new data without having to retrain it on the full training data set.
- If
My_Example_Model
does not exist, the model us saved to it. - If
My_Example_Model
exists and was trained using MLPClassifier, the command updates the existing model with the new input. - If
My_Example_Model
exists but was not trained using MLPClassifier, an error message will be given.
For descriptions of the batch_size
, random_state
and max_iter
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.neural_network.MLPClassifier.html
The hidden_layer_sizes
parameter format (int
) varies based on the number of hidden layers in the data.
You can save MLPClassifier models by using the into
keyword and apply it to new data later by using the apply
command.
You cannot inspect the model learned by the MLPClassifier algorithm with the summary
command.
Using the MLPClassifier algorithm requires running version 1.3 of Python for Scientific Computing.
RandomForestClassifier
The RandomForestClassifier algorithm uses the scikit-learn RandomForestClassifier estimator to fit a model to predict the value of categorical fields.
Syntax
fit RandomForestClassifier <field_to_predict> from <explanatory_fields> [into <model name>] [n_estimators=<int>] [max_depth=<int>] [criterion=<gini | entropy>] [random_state=<int>] [max_features=<str>] [min_samples_split=<int>] [max_leaf_nodes=<int>]
Example
... | fit RandomForestClassifier SLA_violation from * into sla_model | ...
For descriptions of the n_estimators
, max_depth
, criterion
, random_state
, max_features
, min_samples_split
, and max_leaf_nodes
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html.
You can save RandomForestClassifier models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply sla_model
You can list the features that were used to fit the model, as well as their relative importance or influence with the summary
command (for example, | summary sla_model
).
SGDClassifier
The SGDClassifier algorithm uses the scikit-learn SGDClassifier estimator to fit a model to predict the value of categorical fields. This algorithm supports incremental fit.
Syntax
fit SGDClassifier <field_to_predict> from <explanatory_fields> [into <model name>] [partial_fit=<true|false>] [loss=<hinge|log|modified_huber|squared_hinge|perceptron>] [fit_intercept=<true|false>]<int> [random_state=<int>] [n_iter=<int>] [l1_ratio=<float>] [alpha=<float>] [eta0=<float>] [power_t=<float>] [penalty=<l1|l2|elasticnet>] [learning_rate=<constant|optimal|invscaling>]
Example
... | fit SGDClassifier SLA_violation from * into sla_model
The SGDClassifier algorithm supports the following parameters:
loss=<hinge|log|modified_huber|squared_hinge|perceptron>
: The loss function to be used.- Defaults to
hinge
, which gives a linear SVM.
- Defaults to
log
loss gives logistic regression, a probabilistic classifier.modified_huber
is another smooth loss that brings tolerance to outliers as well as probability estimates.squared_hinge
is like hinge but is quadratically penalized.perceptron
is the linear loss used by the perceptron algorithm.fit_intercept=<true|false>
: Specifies whether the intercept should be estimated or not (defaulttrue
).n_iter=<int>
: The number of passes over the training data (aka epochs) (default 5). The number of iterations is set to 1 if using partial_fit.penalty=<l2|l1|elasticnet>
: The penalty (aka regularization term) to be used (default l2).learning_rate=<constant|optimal|invscaling>
The learning rate.constant
: eta = eta0,optimal
: eta = 1.0/(alpha * t),invscaling
: eta = eta0 / pow(t, power_t) (defaultinvscaling
).l1_ratio=<float>
: The Elastic Net mixing parameter, with 0 <= l1_ratio <= 1 (default 0.15). l1_ratio=0 corresponds to L2 penalty, l1_ratio=1 to L1.alpha=<float>
: Constant that multiplies the regularization term (default 0.0001). Also used to compute learning_rate when set tooptimal
.eta0=<float>
: The initial learning rate (default 0.01).power_t=<float>
: The exponent for inverse scaling learning rate (default 0.25).random_state=<int>
: The seed of the pseudo random number generator to use when shuffling the data.
Example
The following example uses the partial_fit=<true|false>
command which specifies whether an existing model should be incrementally updated or not (default setting is false). If My_Incremental_Model
does not exist, the command saves the model data under the model name My_Incremental_Model
. If My_Incremental_Model
exists and was trained using SGDClassifier, the command updates the existing model with the new input. If My_Incremental_Model
exists but was not trained by SGDClassifier, an error message is thrown.
partial_fit | inputlookup iris.csv | fit SGDClassifier species from * partial_fit=true into My_Incremental_Model
Using partial_fit=true
on an existing model ignores the newly supplied parameters. The parameters supplied at model creation are used instead. If partial_fit=false
or partial_fit
is not specified (default is false
), the model specified is created and replaces the pre-trained model if one exists.
You can save SGDClassifier models using the into
keyword and apply the saved model later to new data using the apply
command. See the following example:
Example
... | apply sla_model
You can inspect the model learned by SGDClassifier with the summary
command. See the following example:
Example
| summary sla_model
SVM
The SVM algorithm uses the scikit-learn kernel-based SVC estimator to fit a model to predict the value of categorical fields. It uses the radial basis function (rbf) kernel by default.
Syntax
fit SVM <field_to_predict> from <explanatory_fields> [into <model name>] [C=<float>] [gamma=<float>]
Example
... | fit SVM SLA_violation from * into sla_model | ...
The gamma
parameter controls the width of the rbf kernel (the default value is 1 / number of fields
), and the C
parameter controls the degree of regularization when fitting the model (the default value is 1.0).
For descriptions of the C
and gamma
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html.
You can save SVM models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply sla_model
You cannot inspect the model learned by SVM with the summary
command.
Kernel-based methods such as the scikit-learn SVC tend to work best when the data is scaled, for example, using our StandardScaler algorithm:
| fit StandardScaler
. For details, see ''A Practical Guide to Support Vector Classification'' at https://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf.
Clustering Algorithms
Clustering is the grouping of data points. Results will vary depending upon the clustering algorithm used. Clustering algorithms differ in how they determine if data points are similar, and should be grouped. For example, the K-means algorithm clusters based on points in space, whereas the DBSCAN algorithm clusters based on local density
Birch
The Birch algorithm uses the scikit-learn Birch clustering algorithm to divide data points into set of distinct clusters. The cluster for each event is set in a new field named cluster
. This algorithm supports incremental fit.
Syntax
fit Birch <fields> [into <model name>] [k=<int>][partial_fit=<true|false>] [into <model name>]
Example
... | fit Birch * k=3 | stats count by cluster
- The
k
parameter specifies the number of clusters to divide the data into after the final clustering step, which treats the sub-clusters from the leaves of the CF tree as new samples. By default, the cluster label field name iscluster
. Change that behavior by using theas
keyword to specify a different field name. - The
partial_fit
parameter controls whether an existing model should be incrementally updated or not (default isfalse
). These controls allow you to update an existing model using only new data, without having to retrain the model on the full training data set.
Example
Using partial_fit
:
| inputlookup track_day.csv | fit Birch * k=6 partial_fit=true into My_Incremental_Model
In the example above, if My_Incremental_Model
does not exist, the model is saved to it. If My_Incremental_Model
exists and was trained using Birch, the command updates the existing model with the new input. If My_Incremental_Model
exists but was not trained by Birch, an error message will be given. Using partial_fit=true
on an existing model ignores the newly supplied parameters. The parameters supplied at model creation are used instead. If partial_fit=false
or partial_fit
is not specified (default is false
), the model specified is created and replaces the pre-trained model if one exists.
You can save Birch models using the into
keyword and apply new data later using the apply
command.
Example
... | apply Birch_model
You cannot inspect the model learned by Birch with the summary
command.
DBSCAN
The DBSCAN algorithm uses the scikit-learn DBSCAN clustering algorithm to divide a result set into distinct clusters. The cluster for each event is set in a new field named "cluster". DBSCAN is distinct from K-Means in that it clusters results based on local density, and uncovers a variable number of clusters, whereas K-Means finds a precise number of clusters (for example, k=5 finds 5 clusters).
Syntax
fit DBSCAN <fields> [eps=<number>]
Example
... | fit DBSCAN * | stats count by cluster
The eps
parameter specifies the maximum distance between two samples for them to be considered in the same cluster. By default, the cluster label is assigned to a field named "cluster", but you may change that behavior by using the as
keyword to specify a different field name.
You cannot save DBSCAN models using the into
keyword. If you want to be able to predict cluster assignments for future data, you can combine the DBSCAN algorithm with any clustering algorithm (for example, first cluster the data using DBSCAN, then fit a classifier to predict the cluster using RandomForestClassifier).
K-means
K-means clustering is a type of unsupervised learning. It is a clustering algorithm that groups similar data points, with the number of groups represented by the variable "k". The K-means algorithm uses the scikit-learn K-means implementation. The cluster for each event is set in a new field named cluster
. Use the K-means algorithm when you have unlabeled data and have at least approximate knowledge of the total number of groups into which the data can be divided.
Using the K-means algorithm has the following advantages:
- Computationally faster than most other clustering algorithms.
- Simple algorithm to explain and understand.
- Normally produces tighter clusters than hierarchical clustering.
Using the K-means algorithm has the following disadvantages:
- Difficult to determine optimal or true value of
k
. See X-means. - Sensitive to scaling. See StandardScaler.
- Each clustering may be slightly different, unless you specify the
random_state
parameter. - Does not work well with clusters of different sizes and density.
Syntax
fit KMeans <fields> [into <model name>] [k=<int>] [random_state=<int>]
Example
... | fit KMeans * k=3 | stats count by cluster
The k
parameter specifies the number of clusters to divide the data into. By default, the cluster label is assigned to a field named cluster, but you might change that behavior by using the as
keyword to specify a different field name.
You can save K-means models using the into
keyword when using the fit
command.
Example
In a separate search, you can apply the model to new data using the apply
command.
<code>... | apply cluster_model
Example
You can inspect the model with the summary
command.
... | summary cluster_model
For descriptions of default value of K
, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html
SpectralClustering
The SpectralClustering algorithm uses the scikit-learn SpectralClustering clustering algorithm to divide a result set into set of distinct clusters. SpectralClustering first transforms the input data using the Radial Basis Function (rbf) kernel, and then performs K-Means clustering on the result. Consequently, SpectralClustering can learn clusters with a non-convex shape. The cluster for each event is set in a new field named "cluster".
Syntax
fit SpectralClustering <fields> [k=<int>] [gamma=<float>] [random_state=<int>]
Example
... | fit SpectralClustering * k=3 | stats count by cluster
The k
parameter specifies the number of clusters to divide the data into after kernel step. By default, the cluster label is assigned to a field named "cluster", but you can change that behavior by using the as
keyword to specify a different field name.
You cannot save SpectralClustering models using the into
keyword. If you want to be able to predict cluster assignments for future data, you can combine the SpectralClustering algorithm with any clustering algorithm (for example, first cluster the data using SpectralClustering, then fit a classifier to predict the cluster using RandomForestClassifier).
X-means
Use the X-means algorithm when you have unlabeled data and no prior knowledge of the total number of labels into which that data could be divided. The X-means clustering algorithm is an extended K-means that automatically determines the number of clusters based on Bayesian Information Criterion (BIC) scores. Starting with a single cluster, the X-means algorithm goes into action after each run of K-means, making local decisions about which subset of the current centroids should split themselves in order to fit the data better. The splitting decision is done by computing the BIC. The cluster for each event is set in a new field named cluster, and the total number of clusters is set in a new field named n_clusters
.
Using the X-means algorithm has the following advantages:
- Eliminates the requirement of having to provide the value of
k
(The total number of labels/clusters in the data). - Normally produces tighter clusters than hierarchical clustering.
Using the X-means algorithm has the following disadvantages:
- Sensitive to scaling. See StandardScaler.
- Different initializations might result in different final clusters.
- Does not work well with clusters of different sizes and density.
Syntax
fit XMeans <fields> [into <model name>]
Example
... | fit XMeans * | stats count by cluster
By default, the cluster label is assigned to a field named cluster, but you may change that behavior by using the as
keyword to specify a different field name. You can save X-means models using the into
keyword.
Example
You can then apply new data to the saved X-means model using the apply
command.
... | apply cluster_model
Example
You can inspect the model learned by X-means with the summary
command.
| summary cluster_model
Feature Extraction
Feature extraction algorithms transform fields for better prediction accuracy.
FieldSelector
The FieldSelector algorithm uses the scikit-learn GenericUnivariateSelect to select the best predictor fields based on univariate statistical tests.
Syntax
fit FieldSelector <field_to_predict> from <explanatory_fields> [into <model name>] [type=<categorical, numeric>] [mode=<k_best, fpr, fdr, fwe, percentile>] [param=<int>]
Example
... | fit FieldSelector type=categorical SLA_violation from * into sla_model | ...
The type
parameter specifies if the field to predict is categorical or numeric. For descriptions of the mode
and param
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.GenericUnivariateSelect.html.
You can save FieldSelector models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply sla_model
You cannot inspect the model learned by FieldSelector with the summary
command.
KernelPCA
The KernelPCA algorithm uses the scikit-learn KernelPCA to reduce the number of fields by extracting uncorrelated new features out of data. The difference between KernelPCA and PCA is the use of kernels in the former, which helps with finding nonlinear dependencies among the fields. Currently, KernelPCA only supports the Radial Basis Function (rbf) kernel.
Syntax
fit KernelPCA <fields> [into <model name>] [degree=<int>] [k=<int>] [gamma=<int>] [tolerance=<int>] [max_iteration=<int>]
Example
... | fit KernelPCA * k=3 gamma=0.001 | ...
The k
parameter specifies the number of features to be extracted from the data. The other parameters are for fine tuning of the kernel. For descriptions of the gamma
, degree
, tolerance
, and max_iteration
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.decomposition.KernelPCA.html.
You can save KernelPCA models using the into
keyword and apply new data later using the apply
command. See example below:
Example
... | apply user_feedback_model
You cannot inspect the model learned by KernelPCA with the summary
command.
Kernel-based methods such as the scikit-learn KernelPCA tend to work best when the data is scaled, for example, using our StandardScaler algorithm: | fit StandardScaler
. For details, see ''A Practical Guide to Support Vector Classification'' at https://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf.
PCA
The PCA algorithm uses the scikit-learn PCA algorithm to reduce the number of fields by extracting new uncorrelated features out of the data.
Syntax
fit PCA <fields> [into <model name>] [k=<int>]
Example
... | fit PCA * k=3 | ...
The k
parameter specifies the number of features to be extracted from the data.
You can save PCA models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply user_feedback_model
You cannot inspect the model learned by PCA with the summary
command.
TFIDF
The TFIDF algorithm uses the scikit-learn TfidfVectorizer to convert raw text data into a matrix making it possible to use other machine learning estimators on the data.
Syntax
fit TFIDF <field_to_convert> [into <model name>] [max_features=<int>] [max_df=<int>] [min_df=<int>] [ngram_range=<int>-<int>] [analyzer=<str>] [norm=<str>] [token_pattern=<str>] [stop_words=english]
Example
... | fit TFIDF Reviews into user_feedback_model ngram_range=1-2 max_df=0.6 min_df=0.2 | ...
For descriptions of the max_features
, max_df
, min_df
, ngram_range
, analyzer
, norm
, and token_pattern
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.TfidfVectorizer.html.
The default for max_features
is 100.
To configure the algorithm to ignore common English words (for example, "the", "it", "at", and "that"), set stop_words
to english
. For other languages (for example, machine language) you can ignore the common words by setting max_df
to a value greater than or equal to 0.7 and less than 1.0.
You can save TFIDF models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply user_feedback_model
You cannot inspect the model learned by TFIDF with the summary
command.
Preprocessing
Preprocessing algorithms are used for preparing data. Other algorithms can also be used for preprocessing that may not be organized under this section. For example, PCA can be used for both Feature Extraction and Preprocessing.
RobustScaler
The RobustScaler algorithm uses the scikit-learn RobustScaler algorithm to standardize the data fields by scaling their median and interquartile range to 0 and 1, respectively. It is very similar to the StandardScaler algorithm, in that it helps avoid dominance of one or more fields over others in subsequent machine learning algorithms, and is practically required for some algorithms, such as KernelPCA and SVM. The main difference between StandardScaler and RobustScaler is that RobustScaler is less sensitive to outliers. This algorithm does not support incremental fit.
Syntax
fit RobustScaler <fields> [into <model name>] [with_centering=<true|false>] [with_scaling=<true|false>]
Example
... | fit RobustScaler * | ...
- The
with_centering
andwith_scaling
parameters specify if the fields should be standardized with respect to their median and interquartile range.
You can save RobustScaler models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply scaling_model
You can inspect the statistics extracted by RobustScaler with the summary
command (for example, | summary scaling_model
).
StandardScaler
The StandardScaler algorithm uses the scikit-learn StandardScaler algorithm to standardize data fields by scaling their mean and standard deviation to 0 and 1, respectively. This preprocessing step helps to avoid dominance of one or more fields over others in subsequent machine learning algorithms. This step is practically required for some algorithms, such as KernelPCA and SVM. This algorithm supports incremental fit.
Syntax
fit StandardScaler <fields> [into <model name>] [with_mean=<true|false>] [with_std=<true|false>] [partial_fit=<true|false>]
Example
... | fit StandardScaler * | ...
- The
with_mean
andwith_std
parameters specify if the fields should be standardized with respect to their mean and standard deviation. - The
partial_fit
parameter controls whether an existing model should be incrementally updated or not (default isfalse
). This allows you to update an existing model using only new data without having to retrain it on the full training data set.
Example
The following example uses thepartial_fit
parameter:
| inputlookup track_day.csv | fit StandardScaler "batteryVoltage", "engineCoolantTemperature", "engineSpeed" partial_fit=true into My_Incremental_Model
In the example above, if My_Incremental_Model
does not exist, the model is saved to it. If My_Incremental_Model
exists and was trained using StandardScaler, the command updates the existing model with the new input. If My_Incremental_Model
exists but was not trained by StandardScaler, an error message will be given. Using partial_fit=true
on an existing model ignores the newly supplied parameters. The parameters supplied at model creation are used instead. If partial_fit=false
or partial_fit
is not specified (default is false), the model specified is created and replaces the pre-trained model if one exists.
You can save StandardScaler models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply scaling_model
You can inspect the statistics extracted by StandardScaler with the summary
command. See example below.
Example
| summary scaling_model
Regressors
Regressor algorithms predict the value of a numeric field.
The kfold
cross-validation command is available for all Regressor algorithms. Learn more here.
DecisionTreeRegressor
The DecisionTreeRegressor algorithm uses the scikit-learn DecisionTreeRegressor estimator to fit a model to predict the value of numeric fields.
Syntax
fit DecisionTreeRegressor <field_to_predict> from <explanatory_fields> [into <model_name>] [max_depth=<int>] [max_features=<str>] [min_samples_split=<int>] [random_state=<int>] [max_leaf_nodes=<int>] [splitter=<best|random>]
Example
... | fit DecisionTreeRegressor temperature from date_month date_hour into temperature_model | ...
For descriptions of the max_depth
, random_state
, max_features
, min_samples_split
, max_leaf_nodes
, and splitter
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeRegressor.html.
You can save DecisionTreeRegressor models using the into
keyword and apply it to new data later using the apply
command. See example below.
Example
... | apply model_DTR
You can inspect the decision tree learned by DecisionTreeRegressor with the summary
command (for example, | summary model_DTR
). Furthermore, you can get a JSON representation of the tree by giving json=t
as an argument to the summary
command (for example, | summary model_DTR json=t
). To specify the maximum depth of the tree to summarize, use the limit
argument (for example, | summary model_DTC limit=10
). The default value for the limit
argument is 5.
ElasticNet
The ElasticNet algorithm uses the scikit-learn ElasticNet estimator to fit a model to predict the value of numeric fields. ElasticNet is a linear regression model that includes both L1 and L2 regularization (it is a generalization of Lasso and Ridge).
Syntax
fit ElasticNet <field_to_predict> from <explanatory_fields> [into <model name>] [fit_intercept=<true|false>] [normalize=<true|false>] [alpha=<int>] [l1_ratio=<int>]
Example
... | fit ElasticNet temperature from date_month date_hour normalize=true alpha=0.5 | ...
For descriptions of the fit_intercept
, normalize
, alpha
, and l1_ratio
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.ElasticNet.html.
You can save ElasticNet models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply temperature_model
You can inspect the coefficients learned by ElasticNet with the summary
command. See example below.
Example
| summary temperature_model
GradientBoostingRegressor
This algorithm uses the GradientBoostingRegressor algorithm from scikit-learn to build a regression model by fitting regression trees on the negative gradient of a loss function. For documentation on the parameters, see GradientBoostingRegressor from scikit-learn http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingRegressor.html
Syntax
fit GradientBoostingRegressor <field_to_predict> from <explanatory_fields> [into <model_name>] [loss=<ls|lad|huber|quantile>] [max_features=<str>] [learning_rate=<float>] [min_weight_fraction_leaf=<float>] [alpha=<float>] [subsample=<float>] [n_estimators=<int>] [max_depth=<int>] [min_samples_split=<int>] [min_samples_leaf=<int>] [max_leaf_nodes=<int>] [random_state=<int>]
Example
The following example uses the GradientBoostingRegressor algorithm to fit a model and saves that model as temperature_model
.
... | fit GradientBoostingRegressor temperature from date_month date_hour into temperature_model | ...
Use the apply
method to apply the trained model to the new data.
...|apply temperature_model
To inspect the features learned by GradientBoostingRegressor use the summary
command.
| summary temperature_model
KernelRidge
The KernelRidge algorithm uses the scikit-learn KernelRidge algorithm to fit a model to predict numeric fields. This algorithm uses the radial basis function (rbf) kernel by default.
Syntax
fit KernelRidge <field_to_predict> from <explanatory_fields> [into <model_name>] [gamma=<float>]
Example
... | fit KernelRidge temperature from date_month date_hour into temperature_model | ...
The gamma
parameter controls the width of the rbf kernel (the default value is 1/number of fields). For details, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.kernel_ridge.KernelRidge.html.
You can save KernelRidge models using the into
keyword and apply new data later using the apply
command. See example below.
... | apply sla_model
You cannot inspect the model learned by KernelRidge with the summary
command.
Lasso
The Lasso algorithm uses the scikit-learn Lasso estimator to fit a model to predict the value of numeric fields. Lasso is like LinearRegression, but it uses L1 regularization to learn a linear models with fewer coefficients and smaller coefficients. Lasso models are consequently more robust to noise and resilient against overfitting.
Syntax
fit Lasso <field_to_predict> from <explanatory_fields> [into <model name>] [alpha=<float>]
Example
... | fit Lasso temperature from date_month date_hour | ...
The alpha
parameter controls the degree of L1 regularization. For details, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.Lasso.html.
You can save Lasso models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply temperature_model
You can inspect the coefficients learned by Lasso with the summary
command. See example below.
Example
| summary temperature_model
LinearRegression
The LinearRegression algorithm uses the scikit-learn LinearRegression estimator to fit a model to predict the value of numeric fields.
Syntax
fit LinearRegression <field_to_predict> from <explanatory_fields> [into <model name>] [fit_intercept=<true|false>] [normalize=<true|false>]
Example
... | fit LinearRegression temperature from date_month date_hour into temperature_model | ..
The fit_intercept
parameter specifies whether the model should include an implicit intercept term (the default value is true
).
You can save LinearRegression models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply temperature_model
You can inspect the coefficients learned by LinearRegression with the summary
command. See example below.
Example
| summary temperature_model
RandomForestRegressor
The RandomForestRegressor algorithm uses the scikit-learn RandomForestRegressor estimator to fit a model to predict the value of numeric fields.
Syntax
fit RandomForestRegressor <field_to_predict> from <explanatory_fields> [into <model name>] [n_estimators=<int>] [max_depth=<int>] [random_state=<int>] [max_features=<str>] [min_samples_split=<int>] [max_leaf_nodes=<int>]
Example
... | fit RandomForestRegressor temperature from date_month date_hour into temperature_model | ...
For descriptions of the n_estimators
, random_state
, max_depth
, max_features
, min_samples_split
, and max_leaf_nodes
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html.
You can save RandomForestRegressor models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply temperature_model
You can list the features that were used to fit the model, as well as their relative importance or influence with the summary
command. See example below.
Example
| summary temperature_model
Ridge
The Ridge algorithm uses the scikit-learn Ridge estimator to fit a model to predict the value of numeric fields. Ridge is like LinearRegression, but it uses L2 regularization to learn a linear models with smaller coefficients, making the algorithm more robust to collinearity.
Syntax
fit Ridge <field_to_predict> from <explanatory_fields> [into <model name>] [fit_intercept=<true|false>] [normalize=<true|false>] [alpha=<int>]
Example
... | fit Ridge temperature from date_month date_hour normalize=true alpha=0.5 | ...
The alpha
parameter specifies the degree of regularization (the default value is 1.0). For descriptions of the fit_intercept
, normalize
, and alpha
parameters, see the scikit-learn documentation at http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.Ridge.html.
You can save Ridge models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply temperature_model
You can inspect the coefficients learned by Ridge with the summary
command. See example below.
Example
| summary temperature_model
SGDRegressor
The SGDRegressor algorithm uses the scikit-learn SGDRegressor estimator to fit a model to predict the value of numeric fields. This algorithm supports incremental fit.
Syntax
fit SGDRegressor <field_to_predict> from <explanatory_fields> [into <model name>] [partial_fit=<true|false>] [fit_intercept=<true|false>] [random_state=<int>] [n_iter=<int>] [l1_ratio=<float>] [alpha=<float>] [eta0=<float>] [power_t=<float>] [penalty=<l1|l2|elasticnet>] [learning_rate=<constant|optimal|invscaling>]
Example
... | fit SGDRegressor temperature from date_month date_hour into temperature_model | ...
SGDRegressor supports the following parameters:
partial_fit=<true|false>
: Specifies whether an existing model should be incrementally updated or not (defaultfalse
). See example below.
Example
The following example uses thepartial_fit
parameter.
| inputlookup server_power.csv | fit SGDRegressor "ac_power" from "total-cpu-utilization" "total-disk-accesses" partial_fit=true into My_Incremental_Model
In the example above, if My_Incremental_Model
does not exist, the model is saved to it. If My_Incremental_Model
exists and was trained using SGDRegressor, the command updates the existing model with the new input. If My_Incremental_Model
exists but was not trained by SGDRegressor, an error message will be given. Using partial_fit=true
on an existing model ignores the newly supplied parameters. The parameters supplied at model creation are used instead. If partial_fit=false
or partial_fit
is not specified (default is false
), the model specified is created and replaces the pre-trained model if one exists.
fit_intercept=<true|false>
: Whether the intercept should be estimated or not (defaulttrue
).n_iter=<int>
: The number of passes over the training data (aka epochs) (default 5). The number of iterations is set to 1 if usingpartial_fit
.penalty=<l2|l1|elasticnet>
: The penalty (aka regularization term) to be used (default l2).learning_rate=<constant|optimal|invscaling>
The learning rate. constant: eta = eta0, optimal: eta = 1.0/(alpha * t), invscaling: eta = eta0 / pow(t, power_t) (defaultinvscaling
).l1_ratio=<float>
: The Elastic Net mixing parameter, with 0 <= l1_ratio <= 1 (default 0.15). l1_ratio=0 corresponds to L2 penalty, l1_ratio=1 to L1.alpha=<float>
: Constant that multiplies the regularization term (default 0.0001). Also used to compute learning_rate when set tooptimal
.eta0=<float>
: The initial learning rate (default 0.01).power_t=<float>
: The exponent for inverse scaling learning rate (default 0.25).random_state=<int>
: The seed of the pseudo random number generator to use when shuffling the data.
You can save SGDRegressor models using the into
keyword and apply new data later using the apply
command. See example below.
Example
... | apply temperature_model
You can inspect the coefficients learned by SGDRegressor with the summary
command. See example below.
Example
| summary temperature_model
Time Series Analysis
Time series analysis algorithms provide methods for analyzing time series data in order to extract meaningful statistics and other characteristics of the data and forecast its future values.
ARIMA
The Autoregressive Integrated Moving Average (ARIMA) algorithm uses the StatsModels ARIMA algorithm to fit a model on a time series for better understanding and/or forecasting its future values. An ARIMA model can consist of autoregressive terms, moving average terms, and differencing operations. The autoregressive terms express the dependency of the current value of time series to its previous ones. The moving average terms model the effect of previous forecast errors (also called random shocks or white noise) on the current value. If the time series is non-stationary, differencing operations are used to make it stationary. A stationary process is a stochastic process that its probability distribution does not change over time.
Syntax
fit ARIMA [_time] <field_to_forecast> order=<int>-<int>-<int> [forecast_k=<int>] [conf_interval=<int>] [holdback=<int>]
Example
... | fit ARIMA Voltage order=4-0-1 holdback=10 forecast_k=10
ARIMA requires order
to be specified at fitting time. order
needs three values:
- Number of autoregressive (AR) parameters
- Number of differencing operations (D)
- Number of moving average (MA) Parameters
It also supports the following parameters:
forecast_k=<int>
: Tells ARIMA how many points into the future should be forecasted. If_time
is specified during fitting along with the<field_to_forecast>
, ARIMA will also generate the timestamps for forecasted values. By default,forecast_k
is zero.conf_interval=<1..99>
: This is the confidence interval in percentage around forecasted values. By default it is set to 95%.holdback=<int>
: This is the number of data points held back from the ARIMA model. This can be useful when you want to compare the forecast against known data points. By default, holdback is zero.
Best Practices
- It is highly recommended to send the time series through timechart before sending it into ARIMA to avoid non-uniform sampling time. If
_time
is not to be specified, using timechart is not necessary. - The time series should not have any gaps or missing data otherwise ARIMA will complain. If there are missing samples in the data, using a bigger span in timechart or using streamstats to fill in the gaps with average values can do the trick.
- ARIMA supports one time series at a time.
- ARIMA models cannot be saved and used at a later time in the current version.
- When chaining ARIMA output to another algorithm (i.e. ARIMA itself), keep in mind the length of the data is the length of the original data +
forecast_k
. If you want to maintain theholdback
position, you need to add the number inforecast_k
to yourholdback
value.
See the StatsModels documentation at http://statsmodels.sourceforge.net/devel/generated/statsmodels.tsa.arima_model.ARIMA.html for more information.
Utility Algorithms
Utility algorithms are not machine learning algorithms, but they provide methods to calculate data characteristics. These algorithms facilitate the process of algorithm selection and parameter selection.
Autocorrelation Function
Autocorrelation Function (ACF) calculates the correlation between a sequence and a shifted copy of itself, as a function of shift
. Shift is also referred to as lag or delay.
Syntax
fit ACF <field> [k=<int>] [fft=true|false] [conf_interval=<int>]
Example
... | fit ACF logins k=50 fft=true conf_interval=90
- The
k
parameter specifies the number of lags to return autocorrelation for. By default,k
is 40. - The
fft
parameter specifies whether ACF is computed via Fast Fourier Transform (FFT). By default,fft
is false. - The
conf_interval
parameter specifies the confidence interval in percentage to return. By default,conf_interval
is set to 95.
See the StatsModels documentation at http://www.statsmodels.org/stable/generated/statsmodels.tsa.stattools.acf.html for more information.
Partial Autocorrelation Function
Partial Autocorrelation Function (PACF) gives the partial correlation between a sequence and its lagged values, controlling for the values of lags that are shorter than its own.
Syntax
fit PACF <field> [k=<int>] [method=<ywunbiased|ywmle|ols>] [conf_interval=<int>]
Example
... | fit PACF logins k=20 conf_interval=90
- The
k
parameter specifies the number of lags to return partial autocorrelation for. By default,k
is 40 - The
method
parameter specifies which method for the calculation to use. By default,method
is unbiased. - The
conf_interval
parameter specifies the confidence interval in percentage to return. By default,conf_interval
is set to 95.
See the StatsModels documentation at http://www.statsmodels.org/stable/generated/statsmodels.tsa.stattools.pacf.html for more information.
Using the score command | Import a machine learning algorithm from Splunkbase |
This documentation applies to the following versions of Splunk® Machine Learning Toolkit: 4.0.0
Feedback submitted, thanks!