def __init__(self, ifeature=1):
self.ifeature = ifeature ## =0 explicit feature, =1 kernel
self.icategory = 0 ## =0 vector =number of categories
self.ncategory = 0 ## idf icategory=1 it is the number of categories
self.cat2vector = 0 ## =0 indicator =1 mean 2 =median 3 =tetrahedron
self.mdata = 0
self.itrain = None
self.itest = None
self.dataraw = None
self.data = None ## raw input
self.XTrain = None ## training features
self.XTrainNorm = None ## normalized features
self.XTest = None ## test features
self.XTestNorm = None ## normalized features
self.K = None ## external training kernel
self.Kcross = None ## externel test kernel
self.d1 = None ## norm of left factor of the kernel
self.d2 = None ## norm of right factor of the kernel
self.norm = cls_norm()
self.crossval = cls_crossval()
self.kernel_params = cls_kernel_params()
self.prekernel_params = cls_kernel_params()
self.ioperator_valued = 0
self.title = None
self.kernel_computed = 0
def __init__(self,ifeature=0):
self.ifeature=ifeature ## =0 explicit feature, =1 kernel
self.icategory=0 ## =0 vector =number of categories
self.ncategory=0 ## idf icategory=1 it is the number of categories
self.cat2vector=0 ## =0 indicator =1 mean 2 =median 3 =tetrahedron
self.mdata=0
self.itrain=None
self.itest=None
self.dataraw=None
self.data=None ## raw input
self.XTrain=None ## training features
self.XTrainNorm=None ## normalized features
self.XTest=None ## test features
self.XTestNorm=None ## normalized features
self.Y0=None ## set of distinc feature vectors
self.Y0Norm=None ## set of distinc normalizedfeature vectors
self.K=None ## external training kernel
self.Kcross=None ## externel test kernel
self.Kraw=None
self.Krawcross=None
self.d1=None ## norm of left factor of the kernel
self.d2=None ## norm of right factor of the kernel
## self.ilocal=2
## self.iscale=0
self.norm=cls_norm()
self.crossval=cls_crossval()
self.kernel_params=cls_kernel_params()
self.prekernel_params=cls_kernel_params()
self.title=None
def __init__(self, ifeature=0):
self.ifeature = ifeature ## =0 explicit feature, =1 kernel
self.icategory = 0 ## =0 vector =number of categories
self.ncategory = 0 ## idf icategory=1 it is the number of categories
self.cat2vector = 0 ## =0 indicator =1 mean 2 =median 3 =tetrahedron
self.mdata = 0
self.itrain = None
self.itest = None
self.dataraw = None
self.data = None ## raw input
self.XTrain = None ## training features
self.XTrainNorm = None ## normalized features
self.XTest = None ## test features
self.XTestNorm = None ## normalized features
## self.Y0=None ## set of distinc feature vectors
## self.Y0Norm=None ## set of distinc normalizedfeature vectors
self.K = None ## external training kernel
self.Kcross = None ## externel test kernel
self.d1 = None ## norm of left factor of the kernel
self.d2 = None ## norm of right factor of the kernel
## self.ilocal=2
## self.iscale=0
self.norm = cls_norm()
self.crossval = cls_crossval()
self.kernel_params = cls_kernel_params()
self.prekernel_params = None
self.title = 'mvm_x'
self.xbias = 0.0
return
def __init__(self,ifeature=0):
self.ifeature=ifeature ## =0 explicit feature, =1 kernel
self.icategory=0 ## =0 vector =number of categories
self.ncategory=0 ## idf icategory=1 it is the number of categories
self.cat2vector=0 ## =0 indicator =1 mean 2 =median 3 =tetrahedron
self.mdata=0
self.itrain=None
self.itest=None
self.dataraw=None
self.data=None ## raw input
self.XTrain=None ## training features
self.XTrainNorm=None ## normalized features
self.XTest=None ## test features
self.XTestNorm=None ## normalized features
## self.Y0=None ## set of distinc feature vectors
## self.Y0Norm=None ## set of distinc normalizedfeature vectors
self.K=None ## external training kernel
self.Kcross=None ## externel test kernel
self.Kpre=None ## prekernel can be used to build input and output
self.d1=None ## norm of left factor of the kernel
self.d2=None ## norm of right factor of the kernel
## self.ilocal=2
## self.iscale=0
self.norm=cls_norm()
self.crossval=cls_crossval()
self.kernel_params=cls_kernel_params()
self.prekernel_params=None
self.title='mvm_y'
self.ymax=10.0
self.ymin=-10.0
self.yrange=20
self.ystep=(self.ymax-self.ymin)/self.yrange
self.Y0Tetra=None
self.ndim=4
self.valrange=(0,1,2,3)
self.classweight=np.ones((self.ndim,len(self.valrange)))
