def __init__(self, y, x):
assert len(y) == len(x)
self.prob = prob = svmc.svm_problem()
self.size = size = len(y)
self.y_array = y_array = svmc.new_double(size)
for i in range(size):
svmc.double_setitem(y_array, i, y[i])
self.x_matrix = x_matrix = svmc.svm_node_matrix(size)
data = [None for i in range(size)]
maxlen = 0
for i in range(size):
x_i = x[i]
lx_i = len(x_i)
data[i] = d = seq_to_svm_node(x_i)
svmc.svm_node_matrix_set(x_matrix, i, d)
if isinstance(x_i, dict):
if (lx_i > 0):
maxlen = max(maxlen, max(x_i.keys()))
else:
maxlen = max(maxlen, lx_i)
# bind to instance
self.data = data
self.maxlen = maxlen
prob.l = size
prob.y = y_array
prob.x = x_matrix
def __init__(self, y, x):
assert len(y) == len(x)
self.prob = prob = svmc.svm_problem()
self.size = size = len(y)
self.y_array = y_array = svmc.new_double(size)
for i in xrange(size):
svmc.double_setitem(y_array, i, y[i])
self.x_matrix = x_matrix = svmc.svm_node_matrix(size)
data = [None for i in xrange(size)]
maxlen = 0
for i in xrange(size):
x_i = x[i]
lx_i = len(x_i)
data[i] = d = seq_to_svm_node(x_i)
svmc.svm_node_matrix_set(x_matrix, i, d)
if isinstance(x_i, dict):
if (lx_i > 0):
maxlen = max(maxlen, max(x_i.keys()))
else:
maxlen = max(maxlen, lx_i)
# bind to instance
self.data = data
self.maxlen = maxlen
prob.l = size
prob.y = y_array
prob.x = x_matrix
def predict_values_raw(self, x):
#convert x into SVMNode, allocate a double array for return
n = self.nr_class * (self.nr_class - 1) // 2
data = seq_to_svm_node(x)
dblarr = svmc.new_double(n)
svmc.svm_predict_values(self.model, data, dblarr)
ret = double_array_to_list(dblarr, n)
svmc.delete_double(dblarr)
svmc.svm_node_array_destroy(data)
return ret
def predict_values_raw(self, x):
#convert x into SVMNode, allocate a double array for return
n = self.nr_class*(self.nr_class-1)//2
data = seq_to_svm_node(x)
dblarr = svmc.new_double(n)
svmc.svm_predict_values(self.model, data, dblarr)
ret = double_array_to_list(dblarr, n)
svmc.delete_double(dblarr)
svmc.svm_node_array_destroy(data)
return ret
def predict_probability(self, x):
#c code will do nothing on wrong type, so we have to check ourself
if self.svm_type == NU_SVR or self.svm_type == EPSILON_SVR:
raise TypeError("call get_svr_probability or get_svr_pdf " \
"for probability output of regression")
elif self.svm_type == ONE_CLASS:
raise TypeError("probability not supported yet for one-class " \
"problem")
#only C_SVC, NU_SVC goes in
if not self.probability:
raise TypeError("model does not support probability estimates")
#convert x into SVMNode, alloc a double array to receive probabilities
data = seq_to_svm_node(x)
dblarr = svmc.new_double(self.nr_class)
pred = svmc.svm_predict_probability(self.model, data, dblarr)
pv = double_array_to_list(dblarr, self.nr_class)
svmc.delete_double(dblarr)
svmc.svm_node_array_destroy(data)
p = {}
for i, l in enumerate(self.labels):
p[l] = pv[i]
return pred, p
def predict_probability(self, x):
#c code will do nothing on wrong type, so we have to check ourself
if self.svm_type == NU_SVR or self.svm_type == EPSILON_SVR:
raise TypeError, "call get_svr_probability or get_svr_pdf " \
"for probability output of regression"
elif self.svm_type == ONE_CLASS:
raise TypeError, "probability not supported yet for one-class " \
"problem"
#only C_SVC, NU_SVC goes in
if not self.probability:
raise TypeError, "model does not support probability estimates"
#convert x into SVMNode, alloc a double array to receive probabilities
data = seq_to_svm_node(x)
dblarr = svmc.new_double(self.nr_class)
pred = svmc.svm_predict_probability(self.model, data, dblarr)
pv = double_array_to_list(dblarr, self.nr_class)
svmc.delete_double(dblarr)
svmc.svm_node_array_destroy(data)
p = {}
for i, l in enumerate(self.labels):
p[l] = pv[i]
return pred, p
def double_array(seq):
size = len(seq)
array = svmc.new_double(size)
for i, item in enumerate(seq):
svmc.double_setitem(array, i, item)
return array
def double_array(seq):
size = len(seq)
array = svmc.new_double(size)
for i, item in enumerate(seq):
svmc.double_setitem(array, i, item)
return array