def main() -> int:
"""Train each classifier using all the training data and run on test data."""
y_train, x_train = libsvm_commonutil.svm_read_problem(TRAIN_DATA_PATH)
y_test, x_test = libsvm_commonutil.svm_read_problem(TEST_DATA_PATH)
trained_linear_kernel = libsvm_svmutil.svm_train(y_train, x_train, "-t 0")
_, (train_linear_accuracy, _,
_), _ = libsvm_svmutil.svm_predict(y_train, x_train,
trained_linear_kernel)
_, (test_linear_accuracy, _,
_), _ = libsvm_svmutil.svm_predict(y_test, x_test,
trained_linear_kernel)
trained_polynomial_kernel = libsvm_svmutil.svm_train(
y_train, x_train, "-t 1 -d 5")
_, (train_polynomial_accuracy, _,
_), _ = libsvm_svmutil.svm_predict(y_train, x_train,
trained_polynomial_kernel)
_, (test_polynomial_accuracy, _,
_), _ = libsvm_svmutil.svm_predict(y_test, x_test,
trained_polynomial_kernel)
print(f"Linear Training Accuracy: {train_linear_accuracy}%")
print(f"Linear Test Accuracy: {test_linear_accuracy}%")
print(f"Polynomial Training Accuracy: {train_polynomial_accuracy}%")
print(f"Polynomial Test Accuracy: {test_polynomial_accuracy}%")
return EXIT_CODE_SUCCESS
def TrainSvmLinear(Y, X, sweep_c=range(-2,8)):
num_positives = float(Y.count(1))
num_negatives = float(Y.count(-1))
best_c = -1
best_acc = -1
for c_pow in sweep_c:
current_c = np.power(2.0,c_pow)
prob = svm.svm_problem(Y,X)
param = svm.svm_parameter('-v 5 -t 0 -c %f -w-1 %f -w1 %f -q' % (current_c,
100/num_negatives,
100/num_positives))
current_acc = svm.svm_train(prob, param)
print '%f, %f' % (current_c, current_acc)
if best_acc < current_acc:
best_acc = current_acc
best_c = current_c
# recompute accuracy
param = svm.svm_parameter('-t 0 -c %f -w-1 %f -w1 %f -q' % (best_c,
100/num_negatives,
100/num_positives))
svm_model = svm.svm_train(prob, param)
p_labs, p_acc, p_vals = svm.svm_predict(Y, X, svm_model, '-q')
prob = svm.svm_problem(Y,X)
param = svm.svm_parameter('-t 0 -c %f -w-1 %f -w1 %f -q' % (best_c,
100/num_negatives,
100/num_positives))
svm_model = svm.svm_train(prob, param)
p_labs, p_acc, p_vals = svm.svm_predict(Y, X, svm_model, '-q')
pdb.set_trace()
return svm_model
def part_d():
print("\n--- Part D ---\n")
print("Reading Data")
train_y, train_x = read_data("train")
test_y, test_x = read_data("test")
print("Normalizing")
train_x = normalize(train_x)
test_x = normalize(test_x)
problem = svm_problem(train_y, train_x)
params = "-q -s 0 -t 2 -g 0.05"
results = []
for c in [10**-5, 10**-3, 1, 5, 10]:
c = " -c %f " % c
print("10-fold CV using" + c)
cv_acc = svm_train(problem, params + c + "-v 10")
print("On test data using" + c)
model = svm_train(problem, params + c)
_, test_acc, _ = svm_predict(test_y, test_x, model)
print("C, Accuracy: ", c, cv_acc, test_acc)
results.append((c, cv_acc, test_acc[0]))
def part_c():
print("\n--- Part C ---\n")
print("Reading Data")
train_y, train_x = read_data("train")
test_y, test_x = read_data("test")
print("Normalizing")
train_x = normalize(train_x).tolist()
test_x = normalize(test_x).tolist()
problem = svm_problem(train_y, train_x)
params = svm_parameter("-q -s 0 -c 1")
# Timing calculations
print("Training SVM (linear kernel)")
params.parse_options("-t 0")
model = svm_train(problem, params)
_, p_acc, _ = svm_predict(test_y, test_x, model)
print("Accuracy: ", p_acc)
print("Training SVM (gaussian kernel)")
params.parse_options("-t 2 -g 0.05")
model = svm_train(problem, params)
_, p_acc, _ = svm_predict(test_y, test_x, model)
print("Accuracy: ", p_acc)
def calculate_race():
correct = 0
answers = []
input = []
count = 0
for d in data:
answers.append(question2b_race_truth.truth[count])
input.append(d)
if count == 49:
break
count += 1
prob = svmutil.svm_problem(answers, input)
param = svmutil.svm_parameter('-t 2 -c 4')
param.cross_validation = 1
param.nr_fold = 10
cv = svmutil.svm_train(prob, param)
param = svmutil.svm_parameter('-t 2 -c 4')
m = svmutil.svm_train(prob, param)
count = 0
for d in data:
if count < 50:
count += 1
continue
else:
x0, max_idx = gen_svm_nodearray(d)
p = libsvm.svm_predict(m, x0)
if p == question2b_race_truth.truth[count]:
correct += 1
count += 1
return cv, correct / float(50) * 100
def main() -> int:
"""Use 10-fold cross-validation to predict test accuracy for each kernel."""
y, x = libsvm_commonutil.svm_read_problem(TRAIN_DATA_PATH)
ten_fold_cross_validator = KFold(n_splits=10, shuffle=True)
linear_accuracies = []
polynomial_accuracies = []
for train_index, test_index in ten_fold_cross_validator.split(x):
x_test = np.array(x)[test_index]
x_train = np.array(x)[train_index]
y_test = np.array(y)[test_index]
y_train = np.array(y)[train_index]
# For reference on options used in training, refer to: https://www.csie.ntu.edu.tw/~cjlin/libsvm/
trained_linear_kernel = libsvm_svmutil.svm_train(y_train, x_train, "-t 0")
_, (linear_accuracy, _, _), _ = libsvm_svmutil.svm_predict(y_test, x_test, trained_linear_kernel)
linear_accuracies.append(linear_accuracy)
trained_polynomial_kernel = libsvm_svmutil.svm_train(y_train, x_train, "-t 1 -d 5")
_, (polynomial_accuracy, _, _), _ = libsvm_svmutil.svm_predict(y_test, x_test, trained_polynomial_kernel)
polynomial_accuracies.append(polynomial_accuracy)
print(f"Linear Accuracies: {linear_accuracies}")
print(f"Avg. Linear Accuracy: {sum(linear_accuracies) / len(linear_accuracies)}")
print(f"Polynomial Accuracies: {polynomial_accuracies}")
print(f"Avg. Polynomial Accuracy: {sum(polynomial_accuracies) / len(polynomial_accuracies)}")
return EXIT_CODE_SUCCESS
def TrainSvmRbf(Y, X, sweep_c=range(-5,5), sweep_g=range(-5,5)):
num_negatives = float(Y.count(-1))
num_positives = float(Y.count(1))
best_c = -1
best_g = -1
best_acc = -1
for c_pow in sweep_c:
for g_pow in sweep_g:
current_c = np.power(2.0,c_pow)
current_g = np.power(2.0,g_pow)
prob = svm.svm_problem(Y,X)
param = svm.svm_parameter('-v 5 -t 2 -c %f -g %f -w-1 %f -w1 %f -q' % (current_c,
current_g,
100/num_negatives,
100/num_positives))
current_acc = svm.svm_train(prob, param)
print 'c = %f, g = %f, cv acc = %f' % (current_c, current_g, current_acc)
if best_acc < current_acc:
best_acc = current_acc
best_c = current_c
best_g = current_g
prob = svm.svm_problem(Y,X)
param = svm.svm_parameter('-t 2 -c %f -g %f -w-1 %f -w1 %f -q' % (best_c, best_g,
100/num_negatives,
100/num_positives))
svm_model = svm.svm_train(prob, param)
p_labs, p_acc, p_vals = svm.svm_predict(Y, X, svm_model, '-q')
pdb.set_trace()
return svm_model
def main():
"""Build representation from files."""
parser = argparse.ArgumentParser()
parser.add_argument("-r",
"--regen",
help="increase output verbosity",
action="store_true")
args = parser.parse_args()
if args.regen:
train_class, train_frames = read_dir('dataset/train/*.txt', ' ')
test_class, test_frames = read_dir('dataset/test/*.txt', ' ')
rad_d1 = [get_rad(action) for action in train_frames]
rad_d1_t = [get_rad(action) for action in test_frames]
cust_d1 = [get_custom(action) for action in train_frames]
cust_d1_t = [get_custom(action) for action in test_frames]
with open('rad_d1', 'w') as f:
f.writelines([' '.join(line) + '\n' for line in rad_d1])
with open('rad_d1.t', 'w') as f:
f.writelines([' '.join(line) + '\n' for line in rad_d1_t])
with open('cust_d1', 'w') as f:
f.writelines([' '.join(line) + '\n' for line in cust_d1])
with open('cust_d1.t', 'w') as f:
f.writelines([' '.join(line) + '\n' for line in cust_d1_t])
rad_d2 = reformat(rad_d1, train_class)
rad_d2_t = reformat(rad_d1_t, test_class)
cust_d2 = reformat(cust_d1, train_class)
cust_d2_t = reformat(cust_d1_t, test_class)
with open('rad_d2', 'w') as f:
f.writelines([' '.join(line) + '\n' for line in rad_d2])
with open('rad_d2.t', 'w') as f:
f.writelines([' '.join(line) + '\n' for line in rad_d2_t])
with open('cust_d2', 'w') as f:
f.writelines([' '.join(line) + '\n' for line in cust_d2])
with open('cust_d2.t', 'w') as f:
f.writelines([' '.join(line) + '\n' for line in cust_d2_t])
# Train the models and test with them
y, x = svm.svm_read_problem('rad_d2')
y_t, x_t = svm.svm_read_problem('rad_d2.t')
rad_model = svm.svm_train(y, x, '-s 0 -t 2 -c 2 -g 0.0005')
rad_labels, (rad_acc, *_), _ = svm.svm_predict(y_t, x_t, rad_model)
y, x = svm.svm_read_problem('cust_d2')
y_t, x_t = svm.svm_read_problem('cust_d2.t')
cust_model = svm.svm_train(y, x, '-s 0 -t 2 -c 8 -g 0.0005')
cust_labels, (cust_acc, *_), _ = svm.svm_predict(y_t, x_t, cust_model)
print(f'RAD accuracy: {rad_acc}')
print(f'Custom accuracy: {cust_acc}')
def train(self):
for i in range(4):
self.convert()
#rbf
param1 = svmutil.svm_parameter("-t 2 -b 1 -c 1 -g 0.001")
param2 = svmutil.svm_parameter("-t 2 -b 1 -c 0.1 -g 0.001")
self.mr.append(svmutil.svm_train(self.problem[0], param1))#hist
self.mr.append(svmutil.svm_train(self.problem[1], param2))#vector
#linear
param3 = svmutil.svm_parameter("-t 0 -b 1 -c 0.1")
param4 = svmutil.svm_parameter("-t 0 -b 1 -c 0.01")
self.ml.append(svmutil.svm_train(self.problem[0], param3))#hist
self.ml.append(svmutil.svm_train(self.problem[1], param4))#vector
self.images = self.images[1:]+self.images[:1]
def valid(self,datasets,opt,opp,method = fold,part_ids = None,seed = None,test_data = None):
if seed is None:
# If seed is not set. UNIX time is used as seed.
seed = time.time()
saving_seed = "%s/log/%s.log.seed" % (self._dir,self._name)
with open(saving_seed,"w") as fp:
# Save used seed value.
fp.write("seed:%f\n" % seed)
if part_ids is None:
part_ids = datasets.pids
groups = [(test,train) for test,train in method(part_ids,seed = seed)]
for cnt,pdtsts in enumerate(groups):
# cnt is number of cluster.
if test_data is None:
test = False
ltest,dtest,itest = test2svm_prob(datasets.mkTest(pdtsts[0]))
else:
test = True
ltest,dtest,itest = test2svm_prob(test_data.mkTest(test_data.pids))
print "start %s validation" % (cnt)
ptrn,itrain = train2svm_prob(datasets.mkTrain(pdtsts[1]))
#opt = svm.svm_parameter(opt)
model = svmutil.svm_train(ptrn,opt)
plbl,pacc,pval = svmutil.svm_predict(ltest,dtest,model,opp)
# create saving direcotry
#self._mkdir(cnt)
# create log files
self._save_log(itest,plbl,pval,cnt,test)
model_name = "%s/model/%s.model.%s" % (self._dir,self._name,cnt)
def problem14(X_train, y_train):
label = list(2 * (np.array(y_train) == 0) - 1)
train = np.array(pd.DataFrame(X_train))
for t in range(-3, 2):
c = 10**t
model = svmutil.svm_train(label, X_train, '-g 80 -c %s' % c)
sv_indices = model.get_sv_indices()
def train_list(self):
print('Start to train.')
paras = '-c 4 -t 0 -h 0 -m 1024'
self.y, self.x = self.load_data_list(train=True)
self.model = svmutil.svm_train(self.y, self.x, paras)
svmutil.svm_save_model('./news/svmmodel', self.model)
print('Train finished.')
def test_convert_svmc_linear_raw_multi(self):
iris = load_iris()
X = iris.data[:, :2]
y = iris.target
y[-5:] = 3
prob = svmutil.svm_problem(y, X.tolist())
param = svmutil.svm_parameter()
param.svm_type = SVC
param.kernel_type = svmutil.LINEAR
param.eps = 1
param.probability = 0
if noprint:
param.print_func = noprint
libsvm_model = svmutil.svm_train(prob, param)
node = convert(libsvm_model, "LibSvmNuSvmcMultiRaw",
[('input', FloatTensorType(shape=['None', 2]))])
self.assertTrue(node is not None)
X2 = numpy.vstack([X[:2], X[60:62], X[110:112],
X[147:149]]) # 5x0, 5x1
dump_data_and_model(
X2.astype(numpy.float32),
SkAPICl(libsvm_model),
node,
basename="LibSvmSvmcRaw-Dec3",
verbose=False,
allow_failure=
"StrictVersion(onnxruntime.__version__) <= StrictVersion('0.1.3')")
def test_convert_nusvmc(self):
iris = load_iris()
X = iris.data[:, :2]
y = iris.target
y[y == 2] = 1
prob = svmutil.svm_problem(y, X.tolist())
param = svmutil.svm_parameter()
param.svm_type = NuSVC
param.kernel_type = svmutil.RBF
param.eps = 1
param.probability = 1
if noprint:
param.print_func = noprint
libsvm_model = svmutil.svm_train(prob, param)
node = convert(libsvm_model, "LibSvmNuSvmc",
[('input', FloatTensorType(shape=['None', 'None']))])
self.assertTrue(node is not None)
dump_data_and_model(
X[:5].astype(numpy.float32),
SkAPIClProba2(libsvm_model),
node,
basename="LibSvmNuSvmc-Dec2",
allow_failure=
"StrictVersion(onnxruntime.__version__) <= StrictVersion('0.1.3')")
def kfold(data, labels, k):
try:
import svmutil
except:
return 0
prabs = []
for xxx in range(0, 10):
picks = np.random.choice(len(data), len(data) / k, replace=False)
testLabel = labels[picks]
testPoint = data[picks]
trainPoint = data[np.setdiff1d(range(0, len(data)), picks)]
trainLabel = labels[np.setdiff1d(range(0, len(data)), picks)]
trainLabel = trainLabel.tolist()
trainPoint = trainPoint.tolist()
prob = svmutil.svm_problem(trainLabel, trainPoint)
param = svmutil.svm_parameter('-t 3 -c 4 -b 1 -q')
testLabel = testLabel.tolist()
testPoint = testPoint.tolist()
m = svmutil.svm_train(prob, param)
svmutil.svm_save_model('n.model', m)
p_label, p_acc, p_val = svmutil.svm_predict(testLabel, testPoint, m, '-b 1')
prabs.append(p_acc[0])
print sum(prabs) / float(len(prabs))
print 'std' + str(np.std(prabs))
return sum(prabs) / float(len(prabs))
def svm_predizer(X, Y, X_teste, Y_teste, C, G, K):
"""
Retorna o melhor valor para os parâmetros custo e gamma do SVM radial.
Parâmetros
----------
X : matriz com os dados de treinamento
y : vetor com classes de cada dados de treinamento
X_teste : amostra
Y_teste : classe da amostra
C : lista com valores para custo
G : lista com valores para gamma
K : inteiro indicando o kernel a ser usado
Retorno
-------
resultado : classificação do SVM.
"""
model = svm_train(Y, X, '-c %f -t %d -g %f -q' % (C, K, G))
resultado = svm_predict(Y_teste, X_teste, model)
return resultado
def TrainSvmLinear2(Y, X, sweep_c=range(-2,18)):
num_positives = float(Y.count(1))
num_negatives = float(Y.count(-1))
best_c = -1
best_acc = -1
for c_pow in sweep_c:
current_c = np.power(2.0,c_pow)
param = svm.svm_parameter('-t 0 -c %f -w-1 %f -w1 %f -q' % (current_c,
100/num_negatives,
100/num_positives))
current_pos_acc, current_neg_acc = CrossValidate(Y, X, param)
current_acc = current_pos_acc
print '%f, %f, %f' % (current_c, current_acc, current_neg_acc)
if best_acc < current_acc:
best_acc = current_acc
best_c = current_c
prob = svm.svm_problem(Y,X)
param = svm.svm_parameter('-t 0 -c %f -w-1 %f -w1 %f -q' % (best_c,
100/num_negatives,
100/num_positives))
svm_model = svm.svm_train(prob, param)
p_labs, p_acc, p_vals = svm.svm_predict(Y, X, svm_model, '-q')
return svm_model
def test_default_names(self):
df = pd.DataFrame({'input': self.x})
# Test with probabilities
spec = libsvm.convert(self.libsvm_model).get_spec()
(_, _, probability_lists) = svm_predict(self.y, self.x,
self.libsvm_model, '-b 1 -q')
probability_dicts = [
dict(zip([1, 2], cur_vals)) for cur_vals in probability_lists
]
df['classProbability'] = probability_dicts
metrics = evaluate_classifier_with_probabilities(
spec, df, verbose=False, probabilities='classProbability')
self.assertLess(metrics['max_probability_error'], 0.00001)
# Test model without probabilities
no_probability_model = svmutil.svm_train(self.prob,
svmutil.svm_parameter())
spec = libsvm.convert(no_probability_model).get_spec()
self.assertEqual(len(spec.description.output), 1)
self.assertEqual(spec.description.output[0].name, u'target')
(df['prediction'], _, _) = svm_predict(self.y, self.x,
no_probability_model, ' -q')
metrics = evaluate_classifier(spec, df, verbose=False)
self.assertEquals(metrics['num_errors'], 0)
def _evaluation_test_helper_with_probability(self, labels, allow_slow):
import copy
df = pd.DataFrame(self.x, columns=self.column_names)
y = copy.copy(self.y)
for i, val in enumerate(labels):
y[i] = val
probability_param = '-b 1'
for param1 in self.non_kernel_parameters:
for param2 in self.kernel_parameters:
param_str = ' '.join([self.base_param, param1, param2, probability_param])
# print("PARAMS: ", param_str)
param = svm_parameter(param_str)
model = svm_train(self.prob, param)
# Get predictions with probabilities as dictionaries
(df['prediction'], _, probability_lists) = svm_predict(y, self.x, model, probability_param + ' -q')
probability_dicts = [dict(zip([1, 2], cur_vals)) for cur_vals in probability_lists]
df['probabilities'] = probability_dicts
spec = libsvm.convert(model, self.column_names, 'target', 'probabilities')
if macos_version() >= (10, 13):
metrics = evaluate_classifier_with_probabilities(spec, df, verbose=False)
self.assertEquals(metrics['num_key_mismatch'], 0)
self.assertLess(metrics['max_probability_error'], 0.00001)
if not allow_slow:
break
if not allow_slow:
break
def printSvmValidationAccuracy(input, output): prob = svmutil.svm_problem(output, input) param = getSvmParam(True) accuracy = svmutil.svm_train(prob, param) return accuracy
def train(self,x,y):
"""
training using y=list,x=dict
parameter = string of parameters
"""
prob=su.svm_problem(y,x)
para=""
para+= "-s %d -t %d -d %d -g %f -r %f -c %f -n %f -p %f -e %f -b %d" %\
(
self.type,
self.kernel,
self.degree,
self.gamma,
self.coef0,
self.c,
self.nu,
self.p,
self.eps,
self.prob
)
if(self.v!=0):
para+=" -v %d" % self.v
if(self.q!=0):
para+= " -q"
print para
para1=su.svm_parameter(para)
self.model=su.svm_train(prob,para1)
return True
def test_multi_class_without_probability(self):
# Generate some random data.
# This unit test should not rely on scikit learn for test data.
x, y = [], []
for _ in range(50):
x.append([
random.gauss(200, 30),
random.gauss(-100, 22),
random.gauss(100, 42)
])
y.append(random.choice([1, 2, 10, 12]))
y[0], y[1], y[2], y[3] = 1, 2, 10, 12
column_names = ['x1', 'x2', 'x3']
prob = svmutil.svm_problem(y, x)
df = pd.DataFrame(x, columns=column_names)
for param1 in self.non_kernel_parameters:
for param2 in self.kernel_parameters:
param_str = ' '.join([self.base_param, param1, param2])
param = svm_parameter(param_str)
model = svm_train(prob, param)
# Get predictions with probabilities as dictionaries
(df['prediction'], _, _) = svm_predict(y, x, model, ' -q')
spec = libsvm.convert(model, column_names, 'target')
metrics = evaluate_classifier(spec, df, verbose=False)
self.assertEquals(metrics['num_errors'], 0)
def train_test_model(train_datafile, test_datafile):
from svmutil import svm_read_problem, svm_train, svm_predict
y,x = svm_read_problem(train_datafile)
m = svm_train(y,x,'-t 0 -e .01 -m 1000 -h 0')
y_test,x_test = svm_read_problem(test_datafile)
p_labs,p_acc,p_vals = svm_predict(y_test,x_test,m)
return p_labs, p_acc, p_vals
def setUpClass(self):
"""
Set up the unit test by loading the dataset and training a model.
"""
if not HAS_SKLEARN:
return
if not HAS_LIBSVM:
return
scikit_data = load_boston()
prob = svmutil.svm_problem(
scikit_data['target'] > scikit_data['target'].mean(),
scikit_data['data'].tolist())
param = svmutil.svm_parameter()
param.svm_type = svmutil.C_SVC
param.kernel_type = svmutil.LINEAR
param.eps = 1
libsvm_model = svmutil.svm_train(prob, param)
libsvm_spec = libsvm_converter.convert(libsvm_model,
scikit_data.feature_names,
'target').get_spec()
# Save the data and the model
self.scikit_data = scikit_data
self.libsvm_spec = libsvm_spec
def iqr_model_train(matrix_kernel_train, labels_train, idx2clipid,
svm_para = '-w1 50 -t 4 -b 1 -c 1'):
"""
Light-weighted SVM learning module for online IQR
@param matrix_kernel_train: n-by-n square numpy array with kernel values
between training data
@param labels_train: row-wise labels of training data (1 or True indicates
positive, 0 or False otherwise
@param idx2clipid: idx2clipid(row_idx) returns the clipid for the 0-base row
in matrix
@param svm_para: (optional) SVM learning parameter
@rtype: dictionary with 'clipids_SV': list of clipids for support vectors
@return: output as a dictionary with 'clipids_SV'
"""
log = logging.getLogger('iqr_model_train')
# set training inputs
matrix_kernel_train = np.vstack((np.arange(1, len(matrix_kernel_train)+1),
matrix_kernel_train)).T
log.debug("Done matrix_kernel_train")
problem = svm.svm_problem(labels_train.tolist(), matrix_kernel_train.tolist(), isKernel=True)
log.debug("Done problem")
svm_param = svm.svm_parameter(svm_para)
log.debug("Done svm_param")
# train model
model = svmutil.svm_train(problem, svm_param)
log.debug("Done train model")
# release memory
del problem
del svm_param
log.debug("Done release memory")
# check learning failure
if model.l == 0:
raise Exception('svm model learning failure')
log.debug("Done checking learning failure (no failure)")
n_SVs = model.l
clipids_SVs = []
idxs_train_SVs = svmtools.get_SV_idxs_nonlinear_svm(model)
for i in range(n_SVs):
_idx_1base = idxs_train_SVs[i]
_idx_0base = _idx_1base - 1
clipids_SVs.append(idx2clipid[_idx_0base])
model.SV[i][0].value = i+1 # within SVM model, index needs to be 1-base
log.debug("Done collecting support vector IDs")
#svmutil.svm_save_model(filepath_model, model)
output = dict()
output['model'] = model
output['clipids_SVs'] = clipids_SVs
return output
def classify(train_inputs, train_outputs, test_inputs, test_outputs):
kernel_train, kernel_test = \
compute_kernel_matrices(train_inputs, test_inputs)
counter = 0
average_ap = 0
average_acc = 0
for label in range(train_outputs.shape[1]):
n_tot = train_inputs.shape[0]
n_pos = train_outputs[:, label].sum()
n_neg = n_tot - n_pos
w_pos = np.float32(n_tot)/(2*n_pos)
w_neg = np.float32(n_tot)/(2*n_neg)
option_string = '-t 4 -q -s 0 -b 1 -c %f -w1 %f -w0 %f' % (100, w_pos, w_neg)
model = svmutil.svm_train(
train_outputs[:, label].tolist(),
kernel_train.tolist(),
option_string)
_, accuracy, prob_estimates = svmutil.svm_predict(
test_outputs[:, label].tolist(),
kernel_test.tolist(), model, '-b 1')
ap = compute_aps(np.array(prob_estimates)[:,np.where(np.asarray(model.get_labels())==1)],
test_outputs[:,label])
average_ap += ap
average_acc += accuracy[0]
counter += 1
print 'label = %d, ap = %f, w_neg = %f, w_pos = %f\n' % (
label, ap, w_neg, w_pos)
mean_ap = np.float32(average_ap) / train_outputs.shape[1]
mean_acc = np.float32(average_acc) / train_outputs.shape[1]
print 'mean_ap = %f, mean_acc = %f\n' % (mean_ap, mean_acc)
def main(path, k):
prabs = []
lns = []
for kk in range(0,k-1):
testLabel = []
trainPoint = []
trainLabel = []
testPoint = []
wcCount = 0
for u in os.listdir(path):
if u[-2:] == 'WC':r
wcCount += 1
filePath = path+u
WC = pickle.load(open(filePath, 'rb'))
if wcCount % k == 0 + kk:
testLabel.append(int(u[1]))
testPoint.append(WC)
else:
trainLabel.append(int(u[1]))
trainPoint.append(WC)
lns.append(len(testLabel))
prob = svmutil.svm_problem(trainLabel, trainPoint)
param = svmutil.svm_parameter('-t 0 -c 4 -b 1 -q')
m = svmutil.svm_train(prob, param)
svmutil.svm_save_model('n.model', m)
p_label, p_acc, p_val = svmutil.svm_predict(testLabel, testPoint, m, '-b 1')
prabs.append(p_acc[0])
def CrossValidate(Y, X, param, k_folds=5):
rand_idx = range(len(Y))
random.shuffle(rand_idx)
idx_groups = SplitIntoK(k_folds, rand_idx)
pos_acc = 0
neg_acc = 0
for i in range(k_folds):
test_idx = idx_groups[i]
exclude_test = [idx_groups[j] for j in range(len(idx_groups)) if i != j]
train_idx = list(chain(*exclude_test))
Y_test = [Y[test_i] for test_i in test_idx]
X_test = [X[test_i] for test_i in test_idx]
Y_train = [Y[train_i] for train_i in train_idx]
X_train = [X[train_i] for train_i in train_idx]
# recompute accuracy
prob = svm.svm_problem(Y_train,X_train)
svm_model = svm.svm_train(prob, param)
p_labs, p_acc, p_vals = svm.svm_predict(Y_test, X_test, svm_model, '-q')
tps = sum([1 for j in range(len(p_labs)) if (p_labs[j]==1 and Y_test[j]==1)])
fns = sum([1 for j in range(len(p_labs)) if (p_labs[j]==-1 and Y_test[j]==1)])
tns = sum([1 for j in range(len(p_labs)) if (p_labs[j]==-1 and Y_test[j]==-1)])
fps = sum([1 for j in range(len(p_labs)) if (p_labs[j]==1 and Y_test[j]==-1)])
pos_acc += tps / float(tps + fns)
neg_acc += tns / float(tns + fps)
pos_acc = pos_acc / k_folds
neg_acc = neg_acc / k_folds
return (pos_acc, neg_acc)
def setUpClass(self):
"""
Set up the unit test by loading the dataset and training a model.
"""
if not HAS_LIBSVM:
# setUpClass is still called even if class is skipped.
return
# Generate some random data.
# This unit test should not rely on scikit learn for test data.
self.x, self.y = [], []
random.seed(42)
for _ in range(50):
self.x.append([random.gauss(200, 30), random.gauss(-100, 22)])
self.y.append(random.choice([1, 2]))
self.y[0] = 1 # Make sure 1 is always the first label it sees
self.y[1] = 2
self.column_names = ['x1', 'x2']
self.prob = svmutil.svm_problem(self.y, self.x)
param = svmutil.svm_parameter()
param.svm_type = svmutil.C_SVC
param.kernel_type = svmutil.LINEAR
param.eps = 1
param.probability = 1
self.libsvm_model = svmutil.svm_train(self.prob, param)
def test_convert_svmc(self):
iris = load_iris()
X = iris.data[:, :2]
y = iris.target
y[y == 2] = 1
prob = svmutil.svm_problem(y, X.tolist())
param = svmutil.svm_parameter()
param.svm_type = SVC
param.kernel_type = svmutil.RBF
param.eps = 1
param.probability = 1
if noprint:
param.print_func = noprint
libsvm_model = svmutil.svm_train(prob, param)
node = convert(libsvm_model, "LibSvmSvmc",
[('input', FloatTensorType())])
self.assertTrue(node is not None)
dump_data_and_model(X[:5].astype(numpy.float32),
SkAPIClProba2(libsvm_model),
node,
basename="LibSvmSvmc-Dec2")
def run_cross_validation(dTrain,dTest):
'Work with Polynomal kernel with cross validation'
print '--run_cross_validation--'
print '-- 1 versus 5 with Q = 2 and Cross Validation--'
Cs = [0.0001,0.001,0.01,0.1,1]
Ecvs = [[],[],[],[],[]]
print '-- Train and Test --'
dTrain_shuffle = dTrain
# Try 100 runs with different partitions
for j in range(100):
# roll those dices
shuffle(dTrain_shuffle)
# Get data and formated vectors
dTrain_1vs5 = getDataOneVsOne(dTrain_shuffle,1,5)
X_train_1vs5,Y_train_1vs5 = get_svm_vector_format(dTrain_1vs5)
# Try all Cs with cross validation
for i in range(len(Cs)):
# Type = Polynomial. Degree = 2. Gamma 1.
# Coef = 1. C = Cs[i].Cross Validation at 10. be quiet
options = '-t 1 -d 2 -g 1 -r 1 -c '+str(Cs[i])+ ' -v 10 -q'
m = svm_train(Y_train_1vs5,X_train_1vs5,options)
Ecvs[i].append(100 - m)
# display
print
for i in range(len(Ecvs)):
print 'Ecv = %s \tfor C = %s'%(sum(Ecvs[i])/100.,Cs[i])
print
def train_test():
train_subdir = "data/train/"
test_subdir = "data/test/"
img_kinds = ["happy", "anger", "neutral", "surprise"]
models = {}
params = "-t 0 -c 3"
svm_params = { "happy": params,
"anger": params,
"neutral": params,
"surprise": params}
#train the models
print 'BUILDING TRAIN MODELS'
for img_kind in img_kinds:
print "\t" + img_kind
problem = build_problem(img_kind, train_subdir)
param = svm.svm_parameter(svm_params[img_kind])
models[img_kind] = svmutil.svm_train(problem, param)
print '================================'
#for each image in test set let's see what is the answe
total_count = 0
correct_count = 0
wrong_count = 0
print 'TESTING MODELS'
for img_kind in img_kinds:
images = glob.glob(test_subdir + "f_" + img_kind + "*.jpg")
for image in images:
print "\t" + image
image_data = cv.LoadImage(image)
# Let's see what are the results from the models
results = {}
for kind in img_kinds:
test_data = get_image_features(image_data, True, kind)
predict_input_data = []
predict_input_data.append(test_data)
# do svm query
(val, val_2, label) = svmutil.svm_predict([1] ,predict_input_data, models[kind])
results[kind] = label[0][0]
sorted_results = sorted(results.iteritems(), key=operator.itemgetter(1))
result = sorted_results[len(sorted_results)-1][0]
total_count += 1
if result == img_kind:
print 'YES :' + result
correct_count += 1
else:
print 'NO :' + result
print sorted_results
wrong_count += 1
print '-----------------------'
print '================================'
print "Total Pictures: " + str(total_count)
print "Correct: " + str(correct_count)
print "Wrong: " + str(wrong_count)
print "Accuracy: " + str(correct_count/float(total_count) * 100)
def train(cls, featuresets, params="-t 0 -q"):
"""Train a classifier using the given featuresets.
Args:
featuresets: List of featuresets.
params: Parameter string to pass to svmutil.svm_parameter.
Returns:
SvmClassifier object.
"""
all_features = set()
all_labels = set()
for featuredict, label in featuresets:
all_features.update(set(featuredict.keys()))
all_labels.add(label)
all_labels = sorted(all_labels)
all_features = sorted(all_features)
featureindex = dict(zip(all_features, range(1, len(all_features) + 1)))
labelindex = dict(zip(all_labels, range(1, len(all_labels) + 1)))
vectors, labels = cls.featuresets_to_svm(featureindex, labelindex,
featuresets)
prob = svmutil.svm_problem(labels, vectors)
param = svmutil.svm_parameter(params)
model = svmutil.svm_train(prob, param)
return cls(featureindex, labelindex, model)
def __init__(self,train_feature_file = TRAIN_FEATURE_FILE):
if os.path.exists(SAVED_MODEL):
self.model = svmutil.svm_load_model(SAVED_MODEL)
else:
y, x = svmutil.svm_read_problem(train_feature_file)
self.model = svmutil.svm_train(y, x, '-c 4')
svmutil.svm_save_model(SAVED_MODEL,self.model)
def run(history, tweek, pref, selections):
logstr = ''
#
w, x, y = history.grab(nsample, ntracks, tweek + 1, pref, 180, 90, 80000)
x = xfilter(x)
timer = timer_c()
lsvm = svmutil.svm_train(y, x, options)
#svmutil.svm_save_model('benchmark.svm', lsvm)
#lsvm = svmutil.svm_load_model('benchmark.svm')
print 'svm trained successfully in %s sec with %d samples.' % (str(
float('{0:.3f}'.format(timer.lag()))), len(w))
#
timer = timer_c()
w, x, y = history.grab(1, ntracks, tweek, pref, 180, 90, 80000)
x = xfilter(x)
p_labels, p_acc, p_vals = svmutil.svm_predict(y, x, lsvm, '')
print 'svm predicted successfully in %s sec with %d samples.' % (str(
float('{0:.3f}'.format(timer.lag()))), len(w))
foo = []
for i in xrange(len(y)):
foo.append((p_labels[i], y[i], w[i]))
foo.sort(key=lambda tup: (-tup[0]))
selections['code'].append(foo[0][2])
selections['tweek'].append(tweek)
return [row[1] for row in foo]
def prob20():
import random
gamma = [1, 10, 100, 1000, 10000]
chosen = {1:0, 10:0, 100:0, 1000:0, 10000:0}
for _ in range(100):
Eout = []
for g in gamma:
trainX, testX, trainy, testy = readdat()
mul_label_2_bin(trainy, testy, 0)
trainX = zip(trainX, trainy)
random.shuffle(trainX)
trainX, trainy = zip(*trainX)
valX = trainX[:1000]
valy = trainy[:1000]
trainX = trainX[1000:]
trainy = trainy[1000:]
m = svmutil.svm_train(trainy, trainX, '-s 0 -t 2 -c 0.1 -g %f -h 0'%(g))
p_label, p_acc, p_val = svmutil.svm_predict(valy, valX, m)
Eout.append(100.0 - p_acc[0])
chosen[gamma[Eout.index(min(Eout))]] += 1
print "prob20: ",
for k in chosen.keys():
print "gamma=%d:%d, " % (k, chosen[k]),
print ""
def prob2_to_4():
x = np.array([[1, 0], [0, 1], [0, -1], [-1, 0], [0, 2], [0, -2], [-2, 0]])
y = np.array([-1, -1, -1, 1, 1, 1, 1])
print "===prob 2==="
xf = np.fliplr(x.copy())
print xf * xf - 2 * x + np.array([[3, -3]])
print "===prob 3==="
prob = SVM.svm_problem(y.tolist(), x.tolist())
param = SVM.svm_parameter(
'-t 1 -c 100 -d 2 -r 1 -g 1') # very large C for hard margin
m = SVM.svm_train(prob, param)
sumA = 0
poly = [0] * 6 # xx, xy, yy, x, y, 1
for i in xrange(m.l):
idx = m.sv_indices[i]
alphay = m.sv_coef[0][i]
alpha = abs(m.sv_coef[0][i])
print "{:d} {:+1.2f}".format(idx, alpha)
sumA += alpha
v = x[idx - 1]
poly[0] += alphay * v[0] * v[0]
poly[1] += alphay * v[1] * v[0] * 2
poly[2] += alphay * v[1] * v[1]
poly[3] += alphay * v[0] * 2
poly[4] += alphay * v[1] * 2
# poly[5] += alphay*1 # no need because Sum(alphay) = 0
poly[5] -= m.rho[0]
print "Sum of alpha is {:1.3f}\nb = {}".format(sumA, m.rho[0])
print "{:+2.2f}xx {:+2.2f}xy {:+2.2f}yy {:+2.2f}x {:+2.2f}y {:+2.2f}".format(
*poly)
def test_convert_svmc_raw(self):
iris = load_iris()
X = iris.data[:, :2]
y = iris.target
y[y == 2] = 1
prob = svmutil.svm_problem(y, X.tolist())
param = svmutil.svm_parameter()
param.svm_type = SVC
param.kernel_type = svmutil.RBF
param.eps = 1
param.probability = 0
if noprint:
param.print_func = noprint
libsvm_model = svmutil.svm_train(prob, param)
# known svm runtime dimension error in ONNX Runtime
node = convert(libsvm_model, "LibSvmSvmcRaw",
[('input', FloatTensorType(shape=['None', 'None']))])
self.assertTrue(node is not None)
dump_data_and_model(
X[:5].astype(numpy.float32),
SkAPICl(libsvm_model),
node,
basename="LibSvmSvmcRaw",
allow_failure=
"StrictVersion(onnxruntime.__version__) < StrictVersion('0.5.0')")
def n_gram_svm(class_size, take_size):
cost = np.array([2.0, 2.0, 2.0, 2.0, 2.0])
gamma = np.array([0.0078125, 0.0078125, 0.0078125, 0.0078125, 0.0078125])
tst = time()
preset = np.load(get_feature_file('small'))
vocab_size = int(np.max(preset[:, :-class_size])) + 1
np.random.shuffle(preset)
train_set = preset[take_size:]
train_txt = train_set[:, :-class_size].astype(np.int64)
train_cls = train_set[:, -class_size:].astype(np.float32)
train_dict = [{
gram: 1 if gram in txt else 0
for gram in np.arange(1, vocab_size)
} for txt in train_txt]
train_major: List[Set[int]] = [set() for _ in np.arange(class_size)]
for i, cls in enumerate(train_cls):
for k in np.nonzero(np.abs(cls - np.max(cls)) < 1e-4)[0]:
train_major[k].add(i)
models = []
for k in np.arange(class_size):
problem = svm_problem([
1 if i in train_major[k] else -1 for i in np.arange(len(train_cls))
], train_dict)
param = svm_parameter('-t 0 -c %f -g %f -b 1 -q' % (cost[k], gamma[k]))
models.append(svm_train(problem, param))
train_time = time() - tst
tst = time()
test_set = preset[:take_size]
test_txt = test_set[:, :-5].astype(np.int64)
test_cls = test_set[:, -5:].astype(np.float32)
res = np.array([])
test_dict = [{
gram: 1 if gram in txt else 0
for gram in np.arange(1, vocab_size)
} for txt in test_txt]
for dic, cls in zip(test_dict, test_cls):
prob = np.zeros(class_size)
for k in np.arange(class_size):
_, _, p = svm_predict([], [dic], models[k], '-b 1 -q')
prob[k] = p[0][0]
prob /= np.sum(prob)
res = np.append(
res, cls @ prob / (np.linalg.norm(cls) * np.linalg.norm(prob)))
test_time = time() - tst
test_acc = np.mean(res)
return train_time, test_time, test_acc
def test_default_names(self):
df = pd.DataFrame({"input": self.x})
df["input"] = df["input"].apply(np.array)
# Test with probabilities
spec = libsvm.convert(self.libsvm_model).get_spec()
if _is_macos() and _macos_version() >= (10, 13):
(_, _, probability_lists) = svm_predict(self.y, self.x,
self.libsvm_model,
"-b 1 -q")
probability_dicts = [
dict(zip([1, 2], cur_vals)) for cur_vals in probability_lists
]
df["classProbability"] = probability_dicts
metrics = evaluate_classifier_with_probabilities(
spec, df, verbose=False, probabilities="classProbability")
self.assertLess(metrics["max_probability_error"], 0.00001)
# Test model without probabilities
no_probability_model = svmutil.svm_train(self.prob,
svmutil.svm_parameter())
spec = libsvm.convert(no_probability_model).get_spec()
self.assertEqual(len(spec.description.output), 1)
self.assertEqual(spec.description.output[0].name, u"target")
if _is_macos() and _macos_version() >= (10, 13):
(df["prediction"], _, _) = svm_predict(self.y, self.x,
no_probability_model, " -q")
metrics = evaluate_classifier(spec, df, verbose=False)
self.assertEquals(metrics["num_errors"], 0)
def _test_prob_model(self, param1, param2):
probability_param = '-b 1'
df = self.df
param_str = ' '.join(
[self.base_param, param1, param2, probability_param])
param = svm_parameter(param_str)
model = svm_train(self.prob, param)
# Get predictions with probabilities as dictionaries
(df['prediction'], _,
probability_lists) = svm_predict(self.y, self.x, model,
probability_param + ' -q')
probability_dicts = [
dict(zip([1, 2], cur_vals)) for cur_vals in probability_lists
]
df['probabilities'] = probability_dicts
spec = libsvm.convert(model, self.column_names, 'target',
'probabilities')
if macos_version() >= (10, 13):
metrics = evaluate_classifier_with_probabilities(spec,
df,
verbose=False)
self.assertEquals(metrics['num_key_mismatch'], 0)
self.assertLess(metrics['max_probability_error'], 0.00001)
def Train_SVM_model(PathToFeatureFile):#生成训练模型文件，model.txt
#print(PathToFeatureFile)
y,x=svmutil.svm_read_problem(PathToFeatureFile)
model=svmutil.svm_train(y,x)
modelFilePath=os.path.join(os.path.split(PathToFeatureFile)[0],f"model_{PathToFeatureFile.split('_',1)[1]}")
svmutil.svm_save_model(modelFilePath,model)
print(modelFilePath)
def svm_training_function(vltr, vftr, parameters):
assert isinstance(vltr, list)
assert isinstance(vftr, list)
assert isinstance(parameters, str)
all_parameters = ' -q -s 0 -t 2 ' + parameters
yellow_err('svm_training_function(): parameters = {0}'.format(all_parameters))
return svmutil.svm_train(vltr, vftr, all_parameters)
def _lib_train_libsvm(user_tfidf, num_pos, num_neg, ignore):
sparse_user_tfidf, num_pos, num_neg = _convert_to_sparse_matrix(user_tfidf, num_pos, num_neg, ignore)
labels = ([1] * num_pos) + ([-1] * num_neg)
param = svm_parameter("-t %d" % KERNEL_NUMBER)
prob = svm_problem(labels, sparse_user_tfidf)
modellog = svm_train(prob, param)
return modellog
def trainSVM(kernel, labels):
#need to add an id number as the first column of the list
svmKernel = column_stack((arange(1, len(kernel.tolist()) + 1), kernel))
prob = svm_problem(labels.tolist(), svmKernel.tolist(), isKernel=True)
param = svm_parameter('-t 4')
model = svm_train(prob, param)
return model
def simulate_skin_with_svm(data_size=None, train_params='-s 0 -t 0'):
"""Simulate learning skin data set with libsvm."""
convert_skin_to_svm(data_size)
train_y, train_x = svm.svm_read_problem('skin_train.svm')
model = svm.svm_train(train_y, train_x, train_params)
test_y, test_x = svm.svm_read_problem('skin_test.svm')
p_label, p_acc, p_val = svm.svm_predict(test_y, test_x, model)
def get_cross_val(x, y, x_val, y_val, gamma_c):
prob = svmutil.svm_problem(y, x)
param = svmutil.svm_parameter('-t 2 -q -c {0} -g {1}'.format(gamma_c.C, gamma_c.gamma))
m = svmutil.svm_train(prob, param)
svmutil.svm_save_model("model", m)
p_label_validation, p_acc_validation, p_val_validation = svmutil.svm_predict(y_val, x_val, m)
return p_acc_validation[0]
def trainSVM(trainMatrix, trainCategory):
svm.svm_model.predict = lambda self, x: svm.svm_predict([0], [x], self)[0][0]
prob = svm.svm_problem(trainCategory, trainMatrix)
param = svm.svm_parameter()
param.kernel_type = svm.LINEAR
param.C = 10
model = svm.svm_train(prob, param)
return model
def leave_one_out(y, x, param, n='DUMMY'):
results = []
for i, test in enumerate(zip(y, x)):
training_y = y[:i] + y[i+1:]
training_x = x[:i] + x[i+1:]
problem = svm.svm_problem(training_y, training_x)
model = svmutil.svm_train(problem, param, '-q')
result = svmutil.svm_predict(y[i:i+1], x[i:i+1], model, '-b 1')
results.append(result + (test[0], make_d.decode(x[i], make_d.decode_dic)))
return results
def train_and_get(c, lr):
c.reset_weight()
c.compile_lr(lr)
print '..building the cnn model: %r with lr: %r' % (c.nkerns, lr)
c.fit_lr(n_epochs = 200, slient = True)
features = test_features.tran2libsvm(c.get_feature(x))
m = svm_train(y[:l], features[:l], '-q')
p_label, p_acc, p_val = svm_predict(y[l:], features[l:], m)
ans = p_acc[0]
return ans
def getSVMAccuracy(trainingData):
numOutputs = len(trainingData[0][1])
for outputIndex in range(numOutputs):
inputs = [input for (input, output) in trainingData]
outputs = [output[outputIndex] for (input, output) in trainingData]
prob = svmutil.svm_problem(outputs, inputs)
param = svmAccuracy.getSvmParam(cross_validation_only = True)
model = svmutil.svm_train(prob, param)
print 'output index: %d - %s\n' % (outputIndex, {0 : "Index", 1: "Middle"}[outputIndex])
def _stop_training(self):
super(LibSVMClassifier, self)._stop_training()
self.normalizer = _LabelNormalizer(self.labels)
labels = self.normalizer.normalize(self.labels.tolist())
features = self.data
# Call svm training method.
prob = libsvmutil.svm_problem(labels, features.tolist())
# Train
self.model = libsvmutil.svm_train(prob, self.parameter)
def hiksvm_train(labels, features, beta):
# calculate class prior
np = len([1 for lab in labels if 1 == lab])
nn = len([1 for lab in labels if -1 == lab])
wp = float(beta)/np
wn = (1.0-beta)/nn
wp *= (np+nn)
wn *= (np+nn)
parameters = "-s 0 -c 1 -t %d -w-1 %g -w1 %g" % (KERNEL_TYPE.index("HI"), wn, wp)
model = svm_train(labels, features, parameters)
return model
def train_test_model(train_datafile, test_datafile):
"""
:param train_datafile: relative path
:param test_datafile: relative path
:return: trains a libsvm model using the training data in train_datafile and test in on the data in test_datafile
"""
from svmutil import svm_read_problem, svm_predict, svm_train
y_test, x_test = svm_read_problem(test_datafile)
y_train, x_train = svm_read_problem(train_datafile)
model = svm_train(y_train, x_train, '-t 0 -e .01 -m 1000 -h 0')
p_labs, p_acc, p_vals = svm_predict(y_test, x_test, model)
return p_labs, p_acc, p_vals
def multiclass_train(valid_labels, labels, data, svm_parameters=None):
if svm_parameters == None:
# make default empty parameters
svm_parameters = []
for i in valid_labels:
svm_parameters.append(svmutil.svm_parameter())
models = []
for i in valid_labels:
oaa_labels = relabel_one_against_all(labels, i)
prob = svmutil.svm_problem(oaa_labels, data)
model = svmutil.svm_train(prob, svm_parameters[i])
models.append(model)
return models