3 d@sdZddlmZmZmZddlmZmZmZGdddeZGdddeZ Gd d d eZ Gd d d eZ Gd ddeZ dS)zAFNI's svm interfaces.) TraitedSpectraitsFile) AFNICommandAFNICommandInputSpecAFNICommandOutputSpecc@seZdZejddddZeddddddZed d d d d dZeddddd dZ eddddd dZ eddd,dddZ ej dddZ edddd Zed!d"dd Zejd#d$dZejd%d&dZej d'd(dZejd)d*dZd+S)-SVMTrainInputSpecz#tname: classification or regressionz-type %sT)descargstr mandatoryz1A 3D+t AFNI brik dataset to be used for training.z -trainvol %sF)r r r existscopyfilez %s_vectorsz7output sum of weighted linear support vectors file namez -bucket %sZ_bucketin_file) name_templater r suffixZ name_sourcez%s_modelz.basename for the brik containing the SVM modelz -model %s_modelz %s_alphaszoutput alphas file namez -alpha %sZ_alphasz9byte-format brik file used to mask voxels in the analysisz-mask %sr)r r positionr rz*Flag to enable the omission of a mask filez -nomodelmask)r r zH.1D labels corresponding to the stimulus paradigm for the training data.z-trainlabels %s)r r r zT.1D censor file that allows the user to ignore certain samples in the training data.z -censor %szJstring specifying type of kernel function:linear, polynomial, rbf, sigmoidz -kernel %sz>Specify the maximum number of iterations for the optimization.z-max_iterations %dz-output sum of weighted linear support vectorsz-woutz additional options for SVM-lightz%sN)__name__ __module__ __qualname__rStrZttyperrout_filemodelalphasmaskBoolZ nomodelmaskZ trainlabelsZcensorkernelZIntZmax_iterationsZw_outoptionsr r `_ Examples ======== >>> from nipype.interfaces import afni as afni >>> svmTrain = afni.SVMTrain() >>> svmTrain.inputs.in_file = 'run1+orig' >>> svmTrain.inputs.trainlabels = 'run1_categories.1D' >>> svmTrain.inputs.ttype = 'regression' >>> svmTrain.inputs.mask = 'mask.nii' >>> svmTrain.inputs.model = 'model_run1' >>> svmTrain.inputs.alphas = 'alphas_run1' >>> res = svmTrain.run() # doctest: +SKIP 3dsvmrcstt|j|||S)N)superr# _format_arg)selfnameZ trait_specvalue) __class__r r!r&mszSVMTrain._format_arg) rrr__doc___cmdr input_specr" output_specZ_additional_metadatar& __classcell__r r )r*r!r#Ss r#c@seZdZejddddZedddddZedd d d Zed dd dZ ej dddZ ej dddZ ej dddZ ej dddZejdddZdS)SVMTestInputSpecz=modname is the basename for the brik containing the SVM modelz -model %sT)r r r z6A 3D or 3D+t AFNI brik dataset to be used for testing.z -testvol %s)r r r r z%s_predictionsz$filename for .1D prediction file(s).z-predictions %s)rr r zr*true* class category .1D labels for the test dataset. It is used to calculate the prediction accuracy performancez-testlabels %s)r r r zeFlag to specify that pname files should be integer-valued, corresponding to class category decisions.z -classout)r r zAFlag to prevent writing predicted values for censored time-pointsz-nopredcensordzAFlag to specify that pname files should not be linearly detrendedz -nodetrendz1Specifies multiclass algorithm for classificationz-multiclass %sz additional options for SVM-lightz%sN)rrrrrrrrrZ testlabelsrZclassoutZ nopredcensordZ nodetrendZ multiclassrr r r r!r0qs<r0c@seZdZdZdZeZeZdS)SVMTesta6Temporally predictive modeling with the support vector machine SVM Test Only For complete details, see the `3dsvm Documentation. `_ Examples ======== >>> from nipype.interfaces import afni as afni >>> svmTest = afni.SVMTest() >>> svmTest.inputs.in_file= 'run2+orig' >>> svmTest.inputs.model= 'run1+orig_model' >>> svmTest.inputs.testlabels= 'run2_categories.1D' >>> svmTest.inputs.out_file= 'pred2_model1' >>> res = svmTest.run() # doctest: +SKIP r$N) rrrr+r,r0r-rr.r r r r!r1sr1N) r+baserrrrrrr r"r#r0r1r r r r!sC*