def getKM(self, X):
"""Returns the kernel matrix between the basis vectors and X.
Parameters
----------
X : {array-like, sparse matrix}, shape = [n_samples, n_features]
Returns
-------
K : array, shape = [n_samples, n_bvectors]
kernel matrix
"""
X = array_tools.as_2d_array(X, True)
test_X = X
degree, coef0, gamma = self.degree, self.coef0, self.gamma
if sp.issparse(test_X):
test_X = array_tools.spmat_resize(test_X, self.train_X.shape[1])
else:
test_X = array_tools.as_dense_matrix(test_X)
train_X = self.train_X
K = array_tools.as_array(train_X * test_X.T)
K *= gamma
K += coef0
K = K ** degree
return K.T
def getKM(self, X):
"""Returns the kernel matrix between the basis vectors and X.
Parameters
----------
X: {array-like, sparse matrix}, shape = [n_samples, n_features]
Returns
-------
K : array, shape = [n_samples, n_bvectors]
kernel matrix
"""
test_X = X
degree, coef0, gamma = self.degree, self.coef0, self.gamma
if sp.issparse(test_X):
test_X = array_tools.spmat_resize(test_X, self.train_X.shape[1])
else:
test_X = array_tools.as_dense_matrix(test_X)
train_X = self.train_X
K = array_tools.as_array(train_X * test_X.T)
K *= gamma
K += coef0
K = K**degree
if self.bias != 0:
K += self.bias
return K.T
def __init__(self, W, b = 0.):
self.W = np.squeeze(array_tools.as_array(W))
if self.W.ndim == 0:
self.W = self.W.reshape(1)
#special case: 1-dimensional multi-task predictor
if W.shape[0] == 1 and W.shape[1] > 0:
self.W = self.W.reshape(W.shape[0], W.shape[1])
self.b = np.squeeze(np.array(b))
def __init__(self, W, b=0.):
self.W = np.squeeze(array_tools.as_array(W))
if self.W.ndim == 0:
self.W = self.W.reshape(1)
#special case: 1-dimensional multi-task predictor
if W.shape[0] == 1 and W.shape[1] > 0:
self.W = self.W.reshape(W.shape[0], W.shape[1])
self.b = np.squeeze(np.array(b))
def __init__(self, A, kernel):
self.kernel = kernel
#newbasis = list(set(nonz[0].tolist()))
self.A = A
#if len(newbasis) != A.shape[0]:
# self.A = A.todense()[newbasis]
# newpool = rpool.copy()
# newpool['basis_vectors'] = newbasis
# self.kernel = kernel.__class__.createKernel(**newpool)
self.A = np.squeeze(array_tools.as_array(self.A))
def __init__(self, A, kernel):
self.kernel = kernel
# newbasis = list(set(nonz[0].tolist()))
self.A = A
# if len(newbasis) != A.shape[0]:
# self.A = A.todense()[newbasis]
# newpool = rpool.copy()
# newpool['basis_vectors'] = newbasis
# self.kernel = kernel.__class__.createKernel(**newpool)
self.A = np.squeeze(array_tools.as_array(self.A))
def __init__(self, A, kernel):
self.kernel = kernel
nonz = A.nonzero()
#newbasis = list(set(nonz[0].tolist()))
self.A = A
#if len(newbasis) != A.shape[0]:
# self.A = A.todense()[newbasis]
# newpool = rpool.copy()
# newpool[data_sources.BASIS_VECTORS] = newbasis
# self.kernel = kernel.__class__.createKernel(**newpool)
self.A = array_tools.as_array(self.A)
def __init__(self, W, b=0.):
"""Initializes a primal model
@param W: coefficients of the linear model, one column per task
@type W: numpy matrix
@param b: bias of the model, one column per task
@type b: numpy matrix
"""
#self.W = array_tools.as_dense_matrix(W)
#print type(W), W.shape
self.W = np.squeeze(array_tools.as_array(W))
#print type(self.W), self.W.shape
self.b = np.squeeze(np.array(b))
def __init__(self, W, b=0.0):
"""Initializes a primal model
@param W: coefficients of the linear model, one column per task
@type W: numpy matrix
@param b: bias of the model, one column per task
@type b: numpy matrix
"""
# self.W = array_tools.as_dense_matrix(W)
# print type(W), W.shape
self.W = np.squeeze(array_tools.as_array(W))
# print type(self.W), self.W.shape
self.b = np.squeeze(np.array(b))
def predict(self, X):
"""Computes predictions for test examples.
Parameters
----------
X: {array-like, sparse matrix}, shape = [n_samples, n_features]
test data matrix
Returns
----------
P: array, shape = [n_samples, n_tasks]
predictions
"""
K = self.kernel.getKM(X)
P = np.dot(K, self.A)
P = array_tools.as_array(P)
return P
def getKM(self, X):
"""Returns the kernel matrix between the basis vectors and X.
Parameters
----------
X: {array-like, sparse matrix}, shape = [n_samples, n_features]
Returns
-------
K : array, shape = [n_samples, n_bvectors]
kernel matrix
"""
test_X = X
if sp.issparse(test_X):
test_X = array_tools.spmat_resize(test_X, self.train_X.shape[1])
else:
test_X = array_tools.as_dense_matrix(test_X)
train_X = self.train_X
K = train_X * test_X.T
K = array_tools.as_array(K)
if self.bias != 0:
K += self.bias
return K.T
def getKM(self, X):
"""Returns the kernel matrix between the basis vectors and X.
Parameters
----------
X : {array-like, sparse matrix}, shape = [n_samples, n_features]
Returns
-------
K : array, shape = [n_samples, n_bvectors]
kernel matrix
"""
X = array_tools.as_2d_array(X, True)
test_X = X
if sp.issparse(test_X):
test_X = array_tools.spmat_resize(test_X, self.train_X.shape[1])
else:
test_X = array_tools.as_dense_matrix(test_X)
train_X = self.train_X
K = train_X * test_X.T
K = array_tools.as_array(K)
if self.bias != 0:
K += self.bias
return K.T
def __init__(self, X, regparam=1.0, number_of_clusters=2, kernel='LinearKernel', basis_vectors=None, Y = None, fixed_indices=None, callback=None, **kwargs):
kwargs['X'] = X
kwargs['kernel'] = kernel
if basis_vectors is not None:
kwargs['basis_vectors'] = basis_vectors
self.svdad = adapter.createSVDAdapter(**kwargs)
self.svals = np.mat(self.svdad.svals)
self.svecs = self.svdad.rsvecs
self.regparam = regparam
self.constraint = 0
#if not kwargs.has_key('number_of_clusters'):
# raise Exception("Parameter 'number_of_clusters' must be given.")
self.labelcount = number_of_clusters
if self.labelcount == 2:
self.oneclass = True
else:
self.oneclass = False
self.callbackfun = callback
if Y is not None:
Y_orig = array_tools.as_array(Y)
#if Y_orig.shape[1] == 1:
if len(Y_orig.shape) == 1:
self.Y = np.zeros((Y_orig.shape[0], 2))
self.Y[:, 0] = Y_orig
self.Y[:, 1] = - Y_orig
self.oneclass = True
else:
self.Y = Y_orig.copy()
self.oneclass = False
for i in range(self.Y.shape[0]):
largestind = 0
largestval = self.Y[i, 0]
for j in range(self.Y.shape[1]):
if self.Y[i, j] > largestval:
largestind = j
largestval = self.Y[i, j]
self.Y[i, j] = -1.
self.Y[i, largestind] = 1.
else:
size = self.svecs.shape[0]
ysize = self.labelcount
if self.labelcount is None: self.labelcount = 2
self.Y = RandomLabelSource(size, ysize).readLabels()
self.size = self.Y.shape[0]
self.labelcount = self.Y.shape[1]
self.classvec = - np.ones((self.size), dtype = np.int32)
self.classcounts = np.zeros((self.labelcount), dtype = np.int32)
for i in range(self.size):
clazzind = 0
largestlabel = self.Y[i, 0]
for j in range(self.labelcount):
if self.Y[i, j] > largestlabel:
largestlabel = self.Y[i, j]
clazzind = j
self.classvec[i] = clazzind
self.classcounts[clazzind] = self.classcounts[clazzind] + 1
self.svecs_list = []
for i in range(self.size):
self.svecs_list.append(self.svecs[i].T)
self.fixedindices = []
if fixed_indices is not None:
self.fixedindices = fixed_indices
else:
self.fixedindices = []
self.results = {}
self.solve(self.regparam)
def predictFromPool(self, rpool):
"""Makes real-valued predictions for new examples"""
self.kernel.getKM(rpool[data_sources.PREDICTION_FEATURES])
P = np.dot(K, self.A)
P = array_tools.as_array(P)
return P
def __init__(self, **kwargs):
self.svdad = creators.createSVDAdapter(**kwargs)
self.svals = self.svdad.svals
self.svecs = self.svdad.rsvecs
self.regparam = float(kwargs["regparam"])
self.constraint = 0
#if not kwargs.has_key('number_of_clusters'):
# raise Exception("Parameter 'number_of_clusters' must be given.")
if kwargs.has_key("number_of_clusters"):
self.labelcount = int(kwargs["number_of_clusters"])
else:
self.labelcount = 2
if self.labelcount == 2:
self.oneclass = True
else:
self.oneclass = False
if kwargs.has_key("callback"):
self.callbackfun = kwargs["callback"]
else:
self.callbackfun = None
if kwargs.has_key("train_labels"):
train_labels = kwargs["train_labels"]
else:
train_labels = None
if train_labels != None:
#Y_orig = array_tools.as_labelmatrix(train_labels)
Y_orig = array_tools.as_array(train_labels)
#if Y_orig.shape[1] == 1:
if len(Y_orig.shape) == 1:
self.Y = zeros((Y_orig.shape[0], 2))
self.Y[:, 0] = Y_orig
self.Y[:, 1] = -Y_orig
self.oneclass = True
else:
self.Y = Y_orig.copy()
self.oneclass = False
for i in range(self.Y.shape[0]):
largestind = 0
largestval = self.Y[i, 0]
for j in range(self.Y.shape[1]):
if self.Y[i, j] > largestval:
largestind = j
largestval = self.Y[i, j]
self.Y[i, j] = -1.
self.Y[i, largestind] = 1.
else:
size = self.svecs.shape[0]
ysize = self.labelcount
if self.labelcount == None: self.labelcount = 2
self.Y = RandomLabelSource(size, ysize).readLabels()
self.size = self.Y.shape[0]
self.labelcount = self.Y.shape[1]
self.classvec = -ones((self.size), dtype=int32)
self.classcounts = zeros((self.labelcount), dtype=int32)
for i in range(self.size):
clazzind = 0
largestlabel = self.Y[i, 0]
for j in range(self.labelcount):
if self.Y[i, j] > largestlabel:
largestlabel = self.Y[i, j]
clazzind = j
self.classvec[i] = clazzind
self.classcounts[clazzind] = self.classcounts[clazzind] + 1
self.svecs_list = []
for i in range(self.size):
self.svecs_list.append(self.svecs[i].T)
self.fixedindices = []
if kwargs.has_key("fixed_indices"):
self.fixedindices = kwargs["fixed_indices"]
else:
self.fixedindices = []
self.results = {}
def __init__(self, **kwargs):
self.svdad = creators.createSVDAdapter(**kwargs)
self.svals = self.svdad.svals
self.svecs = self.svdad.rsvecs
self.regparam = float(kwargs["regparam"])
self.constraint = 0
#if not kwargs.has_key('number_of_clusters'):
# raise Exception("Parameter 'number_of_clusters' must be given.")
if kwargs.has_key("number_of_clusters"):
self.labelcount = int(kwargs["number_of_clusters"])
else:
self.labelcount = 2
if self.labelcount == 2:
self.oneclass = True
else:
self.oneclass = False
if kwargs.has_key("callback"):
self.callbackfun = kwargs["callback"]
else:
self.callbackfun = None
if kwargs.has_key("train_labels"):
train_labels = kwargs["train_labels"]
else:
train_labels = None
if train_labels != None:
#Y_orig = array_tools.as_labelmatrix(train_labels)
Y_orig = array_tools.as_array(train_labels)
#if Y_orig.shape[1] == 1:
if len(Y_orig.shape) == 1:
self.Y = zeros((Y_orig.shape[0], 2))
self.Y[:, 0] = Y_orig
self.Y[:, 1] = - Y_orig
self.oneclass = True
else:
self.Y = Y_orig.copy()
self.oneclass = False
for i in range(self.Y.shape[0]):
largestind = 0
largestval = self.Y[i, 0]
for j in range(self.Y.shape[1]):
if self.Y[i, j] > largestval:
largestind = j
largestval = self.Y[i, j]
self.Y[i, j] = -1.
self.Y[i, largestind] = 1.
else:
size = self.svecs.shape[0]
ysize = self.labelcount
if self.labelcount == None: self.labelcount = 2
self.Y = RandomLabelSource(size, ysize).readLabels()
self.size = self.Y.shape[0]
self.labelcount = self.Y.shape[1]
self.classvec = - ones((self.size), dtype = int32)
self.classcounts = zeros((self.labelcount), dtype = int32)
for i in range(self.size):
clazzind = 0
largestlabel = self.Y[i, 0]
for j in range(self.labelcount):
if self.Y[i, j] > largestlabel:
largestlabel = self.Y[i, j]
clazzind = j
self.classvec[i] = clazzind
self.classcounts[clazzind] = self.classcounts[clazzind] + 1
self.svecs_list = []
for i in range(self.size):
self.svecs_list.append(self.svecs[i].T)
self.fixedindices = []
if kwargs.has_key("fixed_indices"):
self.fixedindices = kwargs["fixed_indices"]
else:
self.fixedindices = []
self.results = {}
def __init__(self,
X,
regparam=1.0,
number_of_clusters=2,
kernel='LinearKernel',
basis_vectors=None,
Y=None,
fixed_indices=None,
callback=None,
**kwargs):
kwargs['X'] = X
kwargs['kernel'] = kernel
if basis_vectors != None:
kwargs['basis_vectors'] = basis_vectors
self.svdad = creators.createSVDAdapter(**kwargs)
self.svals = self.svdad.svals
self.svecs = self.svdad.rsvecs
self.regparam = regparam
self.constraint = 0
#if not kwargs.has_key('number_of_clusters'):
# raise Exception("Parameter 'number_of_clusters' must be given.")
self.labelcount = number_of_clusters
if self.labelcount == 2:
self.oneclass = True
else:
self.oneclass = False
self.callbackfun = callback
if Y != None:
Y_orig = array_tools.as_array(Y)
#if Y_orig.shape[1] == 1:
if len(Y_orig.shape) == 1:
self.Y = np.zeros((Y_orig.shape[0], 2))
self.Y[:, 0] = Y_orig
self.Y[:, 1] = -Y_orig
self.oneclass = True
else:
self.Y = Y_orig.copy()
self.oneclass = False
for i in range(self.Y.shape[0]):
largestind = 0
largestval = self.Y[i, 0]
for j in range(self.Y.shape[1]):
if self.Y[i, j] > largestval:
largestind = j
largestval = self.Y[i, j]
self.Y[i, j] = -1.
self.Y[i, largestind] = 1.
else:
size = self.svecs.shape[0]
ysize = self.labelcount
if self.labelcount == None: self.labelcount = 2
self.Y = RandomLabelSource(size, ysize).readLabels()
self.size = self.Y.shape[0]
self.labelcount = self.Y.shape[1]
self.classvec = -np.ones((self.size), dtype=np.int32)
self.classcounts = np.zeros((self.labelcount), dtype=np.int32)
for i in range(self.size):
clazzind = 0
largestlabel = self.Y[i, 0]
for j in range(self.labelcount):
if self.Y[i, j] > largestlabel:
largestlabel = self.Y[i, j]
clazzind = j
self.classvec[i] = clazzind
self.classcounts[clazzind] = self.classcounts[clazzind] + 1
self.svecs_list = []
for i in range(self.size):
self.svecs_list.append(self.svecs[i].T)
self.fixedindices = []
if fixed_indices != None:
self.fixedindices = fixed_indices
else:
self.fixedindices = []
self.results = {}
self.solve(self.regparam)
def __init__(self, A, kernel):
self.kernel = kernel
self.dim = kernel.train_X.shape[1]
self.A = A
self.A = np.squeeze(array_tools.as_array(self.A))
def __init__(self, A, kernel):
self.kernel = kernel
self.dim = kernel.train_X.shape[1]
self.A = A
self.A = np.squeeze(array_tools.as_array(self.A))
def __init__(self, W, b=0.):
self.W = np.squeeze(array_tools.as_array(W))
if self.W.ndim == 0:
self.W = self.W.reshape(1)
self.b = np.squeeze(np.array(b))
