def predict_liblinear(args):
model_name, files = args[0], args[1:]
model = load_model(model_name)
file_objs = [open(fn) for fn in files]
while True:
lines = [f.readline() for f in file_objs]
if '' in lines:
assert len(set(lines)) == 1
break
vector = [float(line.strip()) for line in lines]
xi, max_idx = gen_feature_nodearray(vector)
label = liblinear.predict(model, xi)
#p_label, p_acc, p_val = predict([1], [vector], model)
print class_to_num(label)
def predict_liblinear(args):
model_name, files = args[0], args[1:]
model = load_model(model_name)
file_objs = [open(fn) for fn in files]
while True:
lines = [f.readline() for f in file_objs]
if '' in lines:
assert len(set(lines)) == 1
break
vector = [float(line.strip()) for line in lines]
xi, max_idx = gen_feature_nodearray(vector)
label = liblinear.predict(model, xi)
#p_label, p_acc, p_val = predict([1], [vector], model)
print class_to_num(label)
def predict_one(xi, m):
"""
Return the label and a :class:`c_double` array of decision values of
the test instance *xi* using :class:`LearnerModel` *m*.
*xi* can be a :class:`list` or a :class:`dict` as in LIBLINEAR python
interface. It can also be a LIBLINEAR feature_node array.
.. note::
This function is designed to analyze the result of one instance.
It has a severe efficiency issue and should be used only by
:func:`libshorttext.classifier.predict_single_text`. If many
instances need to be predicted, they should be stored in a file
and predicted by :func:`predict`.
.. warning::
The content of *xi* may be **changed** after the function call.
"""
if isinstance(xi, (list, dict)):
xi = liblinear.gen_feature_nodearray(xi)[0]
elif not isinstance(xi, POINTER(liblinear.feature_node)):
raise TypeError("xi should be a test instance")
print "pppppppppppppppp1"
learner_param = LearnerParameter(m.param_options[0], m.param_options[1])
if m.bias >= 0:
i = 0
while xi[i].index != -1: i += 1
# Already has bias, or bias reserved.
# Actually this statement should be true if
# the data is read by read_SVMProblem.
if i > 0 and xi[i-1].index == m.nr_feature + 1:
i -= 1
xi[i] = liblinear.feature_node(m.nr_feature + 1, m.bias)
xi[i+1] = liblinear.feature_node(-1, 0)
LearnerProblem.normalize_one(xi, learner_param, m.idf)
print "pppppppppppppppp2"
dec_values = (c_double * m.nr_class)()
label = liblinear.liblinear.predict_values(m, xi, dec_values)
return label, dec_values
def predict_one(xi, m):
"""
Return the label and a :class:`c_double` array of decision values of
the test instance *xi* using :class:`LearnerModel` *m*.
*xi* can be a :class:`list` or a :class:`dict` as in LIBLINEAR python
interface. It can also be a LIBLINEAR feature_node array.
.. note::
This function is designed to analyze the result of one instance.
It has a severe efficiency issue and should be used only by
:func:`libshorttext.classifier.predict_single_text`. If many
instances need to be predicted, they should be stored in a file
and predicted by :func:`predict`.
.. warning::
The content of *xi* may be **changed** after the function call.
"""
if isinstance(xi, (list, dict)):
xi = liblinear.gen_feature_nodearray(xi)[0]
elif not isinstance(xi, POINTER(liblinear.feature_node)):
raise TypeError("xi should be a test instance")
learner_param = LearnerParameter(m.param_options[0], m.param_options[1])
if m.bias >= 0:
i = 0
while xi[i].index != -1:
i += 1
# Already has bias, or bias reserved.
# Actually this statement should be true if
# the data is read by read_SVMProblem.
if i > 0 and xi[i - 1].index == m.nr_feature + 1:
i -= 1
xi[i] = liblinear.feature_node(m.nr_feature + 1, m.bias)
xi[i + 1] = liblinear.feature_node(-1, 0)
LearnerProblem.normalize_one(xi, learner_param, m.idf)
dec_values = (c_double * m.nr_class)()
label = liblinear.liblinear.predict_values(m, xi, dec_values)
return label, dec_values