def svm_to_fiksvm(svm_models, weights, feat_dim, params):
num_models = len(weights)
min_vals = params['min_vals']
max_vals = params['max_vals']
nr_bins = params['nr_bins']
c_weights = (c_double * num_models)()
model_ptr_ptr = (POINTER(svm_model) * num_models)()
for t in range(num_models):
#c_weights[t] = c_double(1.0/num_models)
c_weights[t] = c_double(weights[t])
model_ptr_ptr[t] = pointer(svm_models[t])
#print t, svm_models[t].get_nr_class()
c_min_vals = (c_double * feat_dim)()
c_max_vals = (c_double * feat_dim)()
for i in range(feat_dim):
c_min_vals[i] = c_double(min_vals[i])
c_max_vals[i] = c_double(max_vals[i])
new_model = libfiksvm.create_fiksvm_approx_model(model_ptr_ptr, num_models,
c_weights, feat_dim,
c_min_vals, c_max_vals,
nr_bins)
del model_ptr_ptr
del c_weights
del c_min_vals
del c_max_vals
if not new_model:
print("can't do svm_to_fiksvm")
return None
return toPyModel(new_model)
def svm_to_fiksvm(svm_models, weights, feat_dim, params):
num_models = len(weights)
min_vals = params["min_vals"]
max_vals = params["max_vals"]
nr_bins = params["nr_bins"]
c_weights = (c_double * num_models)()
model_ptr_ptr = (POINTER(svm_model) * num_models)()
for t in range(num_models):
# c_weights[t] = c_double(1.0/num_models)
c_weights[t] = c_double(weights[t])
model_ptr_ptr[t] = pointer(svm_models[t])
# print t, svm_models[t].get_nr_class()
c_min_vals = (c_double * feat_dim)()
c_max_vals = (c_double * feat_dim)()
for i in range(feat_dim):
c_min_vals[i] = c_double(min_vals[i])
c_max_vals[i] = c_double(max_vals[i])
new_model = libfiksvm.create_fiksvm_approx_model(
model_ptr_ptr, num_models, c_weights, feat_dim, c_min_vals, c_max_vals, nr_bins
)
del model_ptr_ptr
del c_weights
del c_min_vals
del c_max_vals
if not new_model:
print("can't do svm_to_fiksvm")
return None
return toPyModel(new_model)
def fiksvm_load_model(model_file_name):
"""
fiksvm_load_model(model_file_name) -> model
Load a fik_approx_model from model_file_name and return.
"""
model = libfiksvm.fiksvm_load_model(model_file_name)
if not model:
print("can't open model file %s" % model_file_name)
return None
model = toPyModel(model)
return model
def fiksvm_load_model(model_file_name):
"""
fiksvm_load_model(model_file_name) -> model
Load a fik_approx_model from model_file_name and return.
"""
model = libfiksvm.fiksvm_load_model(model_file_name)
if not model:
print("can't open model file %s" % model_file_name)
return None
model = toPyModel(model)
return model
import svm
labels = [0, 1]
samples = [[0, 0], [0, 1]]
labels = [0, 1, 1, 2]
samples = [[0, 0], [0, 1], [1, 0], [1, 1]]
import svm
labels = [0, 0, 1, 1]
samples = [[1, 1], [1, -1], [-1, 1], [-1, -1]]
param = svm.svm_parameter("-c 1")
problem = svm.svm_problem(labels, samples)
model = svm.libsvm.svm_train(problem, param)
pmodel = svm.toPyModel(model)
pmodel.predict_values(samples[0])
for i in range(len(samples)):
print svm.libsvm.svm_predict(model, svm.gen_svm_nodearray(samples[i])[0])
r = (c_double * 6)()
svm.libsvm.svm_predict_values(model, svm.gen_svm_nodearray(samples[0])[0], r)
import svm
labels = [0, 1]
samples = [[0, 0], [0, 1]]
labels = [0, 1, 1, 2]
samples = [[0, 0], [0, 1], [1, 0], [1, 1]]
import svm
labels = [0, 0, 1, 1]
samples = [[1, 1], [1, -1], [-1, 1], [-1, -1]]
param = svm.svm_parameter('-c 1')
problem = svm.svm_problem(labels, samples)
model = svm.libsvm.svm_train(problem, param)
pmodel = svm.toPyModel(model)
pmodel.predict_values(samples[0])
for i in range(len(samples)):
print svm.libsvm.svm_predict(model, svm.gen_svm_nodearray(samples[i])[0])
r = (c_double * 6)()
svm.libsvm.svm_predict_values(model, svm.gen_svm_nodearray(samples[0])[0], r)
