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 TrainSvmRbf2(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('-t 2 -c %f -g %f -w-1 %f -w1 %f -q' % (current_c,
current_g,
100/num_negatives,
100/num_positives))
current_pos_acc, current_neg_acc = CrossValidate(Y, X, param)
current_acc = current_pos_acc
print 'c = %f, g = %f, cv acc = %f, neg acc = %f' % (current_c, current_g, current_acc,
current_neg_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 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 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 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 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 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 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 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 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 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_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=[1, '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 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=[1, 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_svmc_linear(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.LINEAR
param.eps = 1
param.probability = 1
if noprint:
param.print_func = noprint
libsvm_model = svmutil.svm_train(prob, param)
node = convert(libsvm_model, "LibSvmSvmcLinear",
[('input', FloatTensorType())])
self.assertTrue(node is not None)
dump_data_and_model(
X[:5].astype(numpy.float32),
SkAPIClProba2(libsvm_model),
node,
basename="LibSvmSvmcLinear-Dec2",
allow_failure=
"StrictVersion(onnxruntime.__version__) < StrictVersion('0.5.0')")
def printSvmValidationAccuracy(input, output): prob = svmutil.svm_problem(output, input) param = getSvmParam(True) accuracy = svmutil.svm_train(prob, param) return accuracy
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 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_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 _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 train_fold(self, k, c):
self.print_debug('train_fold', k, c)
folder_name = 'fold_' + str(k) + '/'
file_name = self.filename + '.data.svm'
y, x = svmutil.svm_read_problem(folder_name + file_name)
prob = svmutil.svm_problem(y, x)
param = svmutil.svm_parameter('-s 0 -t 0 -c ' + str(c))
m = svmutil.svm_train(prob, param)
return m
def convert(self):
count = True
self.hist = []
self.vector = []
self.problem = []
true = []
for i in self.images:#all training histogram and vector data together
for j in i.trainDataHist:
self.hist.append(j)
if count == True:
true.append(1)
else:
true.append(0)
for k in i.trainDataVector:
self.vector.append(k)
count = False
self.problem.append(svmutil.svm_problem(true,self.hist))#hist
self.problem.append(svmutil.svm_problem(true,self.vector))#vector
def svm_traing_process(self):
features, labels = self.load_images_dataset(self.training_set_path)
# train a SVM classifier
features = map(list, features)
prob = svmutil.svm_problem(labels, features)
param = svmutil.svm_parameter(self.kernal_option)
model = svmutil.svm_train(prob, param)
res = svmutil.svm_predict(labels, features, model)
return model
def train_fold_polynomial(self, k, c, g, d):
self.print_debug('train_fold_polynomial', k, c, g, d)
folder_name = 'fold_' + str(k) + '/'
file_name = self.filename + '.data.svm'
y, x = svmutil.svm_read_problem(folder_name + file_name)
prob = svmutil.svm_problem(y, x, isKernel=True)
param = svmutil.svm_parameter('-s 0 -t 1 -c ' + str(c)
+ ' -g ' + str(g) + ' -d ' + str(d))
m = svmutil.svm_train(prob, param)
return m
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 __init__(self, classified_data_list, kernel):
self.kernel = kernel
data = np.empty([0, classified_data_list[0].shape[1]])
labels = np.empty([0])
for i, d in enumerate(classified_data_list):
data = np.vstack((data, d))
labels = np.append(labels, i * np.ones(d.shape[0]))
problem = svmutil.svm_problem(labels.tolist(), data.tolist())
kmap = {'poly': 1, 'rbf': 2, 'sigmoid': 3}
param = svmutil.svm_parameter('-q -c {} -t {}'.format(C, kmap[kernel]))
self.svm = svmutil.svm_train(problem, param)
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 _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 train_grasp(grasp_type, side):
"""
train_grasp(grasp_type):
train linear svm classifier for specific grasp type\n
grasp_type: hand grasping type\n
side: left hand or right hand\n
"""
#train
datafile = "model/traindata_grasp_" + grasp_type + "_" + side
if not os.path.isfile(datafile):
srcfile = "data/feature_grasp_train.csv"
write_svmdata_grasp(srcfile, datafile, grasp_type, side, 0)
label_train, data_train = svmutil.svm_read_problem(datafile)
modelfile = "model/model_grasp_" + grasp_type + "_" + side
m = []
if not os.path.isfile(modelfile):
print("train model: " + grasp_type + "_" + side)
label_weight = {}
for v in label_train:
if label_weight.has_key(v):
label_weight[v] += 1
else:
label_weight[v] = 1
sorted_label = sorted(label_weight)
param_weight = ' '
for v in sorted_label:
label_weight[v] = float(
len(label_train)) / len(sorted_label) / label_weight[v]
param_weight += '-w%d %f ' % (v, label_weight[v])
prob = svmutil.svm_problem(label_train, data_train)
param = svmutil.svm_parameter('-t 0 -b 1 -q' + param_weight)
print '-t 0 -b 1 -q' + param_weight
# param = svmutil.svm_parameter('-t 0 -c 4 -b 1 -q')
m = svmutil.svm_train(prob, param)
svmutil.svm_save_model(modelfile, m)
else:
print("load model: " + grasp_type + "_" + side)
m = svmutil.svm_load_model(modelfile)
#test
grasp_info = read_info("data/feature_grasp_test.csv", side)
datafile = "model/testdata_grasp_" + grasp_type + "_" + side
if not os.path.isfile(datafile):
srcfile = "data/feature_grasp_test.csv"
write_svmdata_grasp(srcfile, datafile, grasp_type, side, 1)
label_test, data_test = svmutil.svm_read_problem(datafile)
p_label, p_acc, p_val = svmutil.svm_predict(label_test, data_test, m,
'-b 1')
f_result = open("result/grasp_" + grasp_type + "_" + side + ".csv", "w")
for i in range(len(p_label)):
f_result.write(grasp_info[i] + ", " + str(int(label_test[i])) + ", " +
str(int(p_label[i])) + ", ")
f_result.write("[%.4f]\n" % p_val[i][0])
f_result.close()
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
def __init__(self, data=[], labels=[], kernel=svmutil.RBF, c=10):
self.__svmparam__ = svmutil.svm_parameter('-q')
self.__svmparam__.kernel_type = kernel
self.__svmparam__.C = c
self.c = c
self.data = data
self.labels = labels
if len(data) > 0:
self.problem = svmutil.svm_problem(labels,data)
self.model = svmutil.svm_train(self.problem,self.__svmparam__,'-q')
else:
self.problem = None
self.model = None
def SVM(X, Y, gamma):
X = X.tolist()
Y = Y.tolist()
prob = svm_problem(Y, X)
param = svm_parameter('-g ' + str(gamma) + ' -t 2 -c 1e6') # hard margin
m = svm_train(prob, param)
# return model
def g(X):
N = X.shape[0]
Y = np.zeros(N)
return np.array(
svm_predict(Y.tolist(), X.tolist(), m, options='-q')[0])
return g
def train_manipulation(mnp_type):
"""
"""
#train
datafile = "manipulate/model/traindata_mnp_"+mnp_type
if not os.path.isfile(datafile):
srcfile = "manipulate/data/feature_mnp_train.csv"
write_svmdata_mnp(srcfile, datafile, mnp_type, 0)
label_train,data_train = svmutil.svm_read_problem(datafile)
modelfile = "manipulate/model/model_mnp_"+mnp_type
m = []
if not os.path.isfile(modelfile):
print("train model: " + mnp_type)
label_weight = {}
for v in label_train:
if label_weight.has_key(v):
label_weight[v]+=1
else:
label_weight[v]=1
sorted_label = sorted(label_weight)
param_weight = ' '
for v in sorted_label:
label_weight[v] = float(len(label_train))/len(sorted_label)/label_weight[v]
param_weight += '-w%d %f ' % (v, label_weight[v])
prob = svmutil.svm_problem(label_train, data_train)
param = svmutil.svm_parameter('-t 0 -b 1 -q'+param_weight)
print '-t 0 -b 1 -q'+param_weight
m = svmutil.svm_train(prob, param)
svmutil.svm_save_model(modelfile, m)
else:
print("load model: " + mnp_type)
m = svmutil.svm_load_model(modelfile)
# weight = read_model_linearSVM(modelfile, len(data_train[0]))
# print weight
#test
mnp_info = read_info("manipulate/data/feature_mnp_test.csv")
datafile = "manipulate/model/testdata_mnp_"+mnp_type
if not os.path.isfile(datafile):
srcfile = "manipulate/data/feature_mnp_test.csv"
write_svmdata_mnp(srcfile, datafile, mnp_type, 1)
label_test,data_test = svmutil.svm_read_problem(datafile)
p_label, p_acc, p_val = svmutil.svm_predict(label_test, data_test, m, '-b 1')
f_result = open("manipulate/result/mnp_" + mnp_type + ".csv", "w")
for i in range(len(p_label)):
f_result.write(mnp_info[i]+", "+str(int(label_test[i]))+", "+str(int(p_label[i]))+", ")
f_result.write("[%.4f]\n" % p_val[i][0])
f_result.close()
def run_svm_model(dbCur,trainY,trainX,testY,testX,time_stamp,round,num_cutoff_points):
trainY = (2*trainY-1).tolist(); testYSVM = (2*testY-1).tolist() # Convert labels to -1,+1 format and lists as required by libsvm.
trainY = [i[0] for i in trainY]; testYSVM = [i[0] for i in testY] # Convert to list of floats from list of lists.
trainX = trainX.tolist(); testX = testX.tolist() # Convert to list as required by libsvm.
prob = svm.svm_problem(trainY,trainX)
params = svm.svm_parameter('-b 1 -q')
svmmodel = svm.svm_train(prob,params)
p_label, p_acc, p_val = svm.svm_predict(testYSVM,testX,svmmodel,'-b 1')
probs = n.array(p_val)[:,1]; probs.shape = (len(probs),1)
performance_results,valid_ind = m.compute_pred_performance_curve(simple_comp,testY,probs,None,num_cutoff_points)
for idx in range(num_cutoff_points):
if valid_ind[0,idx] == 1:
recall = performance_results[0,idx]; precision = performance_results[1,idx]
else:
recall = None; precision = None
dbCur.execute("INSERT INTO snp_results_detail (time_stamp,model_used,fold_no,seq_no,val_recall,val_precision) \
VALUES(timestamp %s,'svm',%s,%s,%s,%s);",[time_stamp,round+1,idx,recall,precision])
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['data'].tolist())
param = svmutil.svm_parameter()
param.svm_type = svmutil.NU_SVR
param.kernel_type = svmutil.LINEAR
param.eps = 1
self.libsvm_model = svmutil.svm_train(prob, param)
def train(self, learning_set):
"""Train the classifier with a list of character objects that have
known values."""
classes = []
features = []
l = len(learning_set)
for i, char in enumerate(learning_set):
if self.verbose:
print 'Found "%s" -- %d of %d (%d%% done)' % (char.value, i + 1, l, round(100 * (i + 1) / l))
classes.append(float(ord(char.value)))
# features.append(char.get_feature_vector())
char.get_single_cell_feature_vector(self.neighbours)
features.append(char.feature)
problem = svm_problem(classes, features)
self.model = svm_train(problem, self.param)
def train(self, learning_set):
"""Train the classifier with a list of character objects that have
known values."""
classes = []
features = []
l = len(learning_set)
for i, char in enumerate(learning_set):
if self.verbose:
print 'Found "%s" -- %d of %d (%d%% done)' \
% (char.value, i + 1, l, round(100 * (i + 1) / l))
classes.append(float(ord(char.value)))
#features.append(char.get_feature_vector())
char.get_single_cell_feature_vector(self.neighbours)
features.append(char.feature)
problem = svm_problem(classes, features)
self.model = svm_train(problem, self.param)
def cSvmTrainSet(self):
dataMat = []
labelMat = []
file_pattern = re.compile('^%s-\d.rec' % self.legalName)
for fdata in os.listdir('data'):
if file_pattern.match(fdata):
data,label = loadDataSet('data/'+fdata,1)
else:
data,label = loadDataSet('data/'+fdata,-1)
dataMat+=data
labelMat+=label
libSvmFormatSaveInFile(dataMat,labelMat,'data_format/%s.mat' % self.legalName) # todo: duoxiancheng
y,x = svmutil.svm_read_problem('data_format/%s.mat' % self.legalName)
prob = svmutil.svm_problem(y,x,isKernel = True)
param = svmutil.svm_parameter('-t 0 ')
self.model = svmutil.svm_train(prob,param)
print self.model
def cSvmTrainSet(self):
dataMat = []
labelMat = []
file_pattern = re.compile('^%s-\d.rec' % self.legalName)
for fdata in os.listdir('data'):
if file_pattern.match(fdata):
data, label = loadDataSet('data/' + fdata, 1)
else:
data, label = loadDataSet('data/' + fdata, -1)
dataMat += data
labelMat += label
libSvmFormatSaveInFile(dataMat, labelMat, 'data_format/%s.mat' %
self.legalName) # todo: duoxiancheng
y, x = svmutil.svm_read_problem('data_format/%s.mat' % self.legalName)
prob = svmutil.svm_problem(y, x, isKernel=True)
param = svmutil.svm_parameter('-t 0 ')
self.model = svmutil.svm_train(prob, param)
print self.model
def _evaluation_test_helper_no_probability(self, labels, allow_slow):
# Generate some random data.
# This unit test should not rely on scikit learn for test data.
x, y = [], []
random.seed(42)
for _ in range(50):
x.append([
random.gauss(200, 30),
random.gauss(-100, 22),
random.gauss(100, 42)
])
y.append(random.choice(labels))
# make sure first label is seen first, second is seen second, and so on.
for i, val in enumerate(labels):
y[i] = val
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])
print("PARAMS: ", param_str)
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")
if _is_macos() and _macos_version() >= (10, 13):
metrics = evaluate_classifier(spec, df, verbose=False)
self.assertEquals(metrics["num_errors"], 0)
if not allow_slow:
break
if not allow_slow:
break
def test_convert_svmr_linear(self):
iris = load_iris()
X = iris.data[:, :2]
y = iris.target
prob = svmutil.svm_problem(y, X.tolist())
param = svmutil.svm_parameter()
param.svm_type = SVR
param.kernel_type = svmutil.LINEAR
param.eps = 1
if noprint:
param.print_func = noprint
libsvm_model = svmutil.svm_train(prob, param)
node = convert(libsvm_model, "LibSvmSvmrLinear", [('input', FloatTensorType(shape=[1, 'None']))])
self.assertTrue(node is not None)
dump_data_and_model(X[:5].astype(numpy.float32), SkAPIReg(libsvm_model), node,
basename="LibSvmSvmrLinear-Dec3")
def main():
parser = argparse.ArgumentParser()
parser.add_argument('-r', '--resdir', type=str, required=True,
help="Results directory")
parser.add_argument('-f', '--feature', type=str, required=True,
help='feature to use to learn')
args = parser.parse_args()
SCORES_FPATH = os.path.join(args.resdir, 'scores.txt')
FEAT_DIR = os.path.join(args.resdir, 'features', args.feature)
scores = np.fromfile(SCORES_FPATH, sep='\n')
feats = []
for i in range(1, len(scores) + 1):
feats.append(np.fromfile(os.path.join(FEAT_DIR, str(i) + '.txt'), sep='\n').tolist())
# feats = np.array(feats)
print('Read all features')
params = svmutil.svm_parameter('-s 4 -t 2')
model = svmutil.svm_train(svmutil.svm_problem(scores, feats), params)
svmutil.svm_save_model(os.path.join(args.resdir, 'svr.model'), model)
print svmutil.svm_predict(scores, feats, model)
def compute_auc(gram_matrix, data, k=10, C=1.0):
kv = cross_validation.StratifiedKFold(labels, n_folds=k)
s = 0.0
for train_index, test_index in kv:
gm_train = gram_matrix[train_index, :]
gm_train = gm_train[:, train_index]
data_train = data[train_index]
# libSVM wants the distances from test instances to all train instances as input
# see http://stackoverflow.com/questions/10978261/libsvm-precomputed-kernels
gm_test = gram_matrix[test_index, :]
gm_test = gm_test[:, train_index] #!
data_test = data[test_index]
# Have to use libsvm directly here, because of a bug in sklearn with precomputed gram matrices
x = []
for i in range(len(gm_train)):
l = gm_train[i].tolist()
l.insert(0, i + 1)
x.append(l)
prob = svmutil.svm_problem(data_train.tolist(), x, isKernel=True)
param = svmutil.svm_parameter("-t 4 -c %.410f -q" % C)
m = svmutil.svm_train(prob, param)
xx = []
for i in range(len(gm_test)):
t = gm_test[i].tolist()
t.insert(0, i + 1)
xx.append(t)
p_label, p_acc, p_val = svmutil.svm_predict(data_test.tolist(), xx, m)
fpr, tpr, thresholds = roc_curve(data_test, p_val, pos_label=1.0)
AUC = roc_auc_score(data_test, p_val)
s += AUC
return s / k
def main(path):
label = []
points = []
for u in os.listdir(path):
if u[-2:] == 'WC':
filePath = path+u
WC = pickle.load(open(filePath, 'rb'))
label.append(u[1])
points.append(WC)
label = [int(i) for i in label]
prob = svmutil.svm_problem(label, points)
param = svmutil.svm_parameter('-t 0 -c 4 -b 1')
m = svmutil.svm_train(prob, param)
svmutil.svm_save_model('n.model', m)
p_label, p_acc, p_val = svmutil.svm_predict(label, points, m, '-b 1')
return p_acc
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 train_attribute(attribute, side):
"""
train_attribute(str, float):
train linear svm classifier for specific attribute\n
attribute: should be one from ["prismatic", "sphere", "flat", "rigid"]
"""
#train
datafile = "model/traindata_attribute_"+attribute+"_"+side
if not os.path.isfile(datafile):
srcfile = "data/feature_attribute_train.csv"
write_svmdata_attribute(srcfile, datafile, attribute, side, 0)
label_train,data_train = svmutil.svm_read_problem(datafile)
modelfile = "model/model_attribute_"+attribute+"_"+side
m = []
if not os.path.isfile(modelfile):
print("train model: " + attribute+"_"+side)
prob = svmutil.svm_problem(label_train, data_train)
param = svmutil.svm_parameter('-t 0 -c 4 -b 1 -q')
m = svmutil.svm_train(prob, param)
svmutil.svm_save_model(modelfile, m)
else:
print("load model: " + attribute+"_"+side)
m = svmutil.svm_load_model(modelfile)
#test
attribute_info = read_info("data/feature_attribute_test.csv", side)
datafile = "model/testdata_attribute_"+attribute+"_"+side
if not os.path.isfile(datafile):
srcfile = "data/feature_attribute_test.csv"
write_svmdata_attribute(srcfile, datafile, attribute, side, 1)
label_test,data_test = svmutil.svm_read_problem(datafile)
p_label, p_acc, p_val = svmutil.svm_predict(label_test, data_test, m, '-b 1')
[precision, recall, f1, accuracy] = getF1(label_test, p_label)
print "F1: [%.4f, %.4f, %.4f] Accuracy: %.4f" % (precision, recall, f1, accuracy)
f_result = open("result/attribute_"+attribute+"_"+side+".csv", "w")
for i in range(len(p_label)):
f_result.write(attribute_info[i]+", "+str(int(label_test[i]))+", "+str(int(p_label[i]))+", ")
f_result.write("[%.4f]\n" % p_val[i][0])
f_result.close()
def create_svm_problem(feats, labelledRects):
labels = []
values = []
height, width, nd = feats.shape
for rect in labelledRects:
x0, y0, x1, y1 = rect.box
if x0 >= width or y0 >= height:
continue
if x1 >= width:
x1 = width - 1
if y1 >= height:
y1 = height - 1
for x in range(x0, x1 + 1):
for y in range(y0, y1 + 1):
vs = []
for i in feats[y][x]:
vs.append(i)
labels.append(rect.label)
values.append(vs)
return svmutil.svm_problem(labels, values)
def bench_svm(X, Y):
"""
bench with swig-generated wrappers that come with libsvm
"""
import svmutil
X1 = X.tolist()
Y1 = Y.tolist()
gc.collect()
# start time
tstart = datetime.now()
problem = svmutil.svm_problem(Y1, X1)
param = svmutil.svm_parameter()
param.svm_type=0
param.kernel_type=2
model = svmutil.svm_train(problem, param)
svmutil.svm_predict([0]*len(X1), X1, model)
delta = (datetime.now() - tstart)
# stop time
svm_results.append(delta.seconds + delta.microseconds/mu_second)
def bench_svm(X, Y):
"""
bench with swig-generated wrappers that come with libsvm
"""
import svmutil
X1 = X.tolist()
Y1 = Y.tolist()
gc.collect()
# start time
tstart = datetime.now()
problem = svmutil.svm_problem(Y1, X1)
param = svmutil.svm_parameter()
param.svm_type = 0
param.kernel_type = 2
model = svmutil.svm_train(problem, param)
svmutil.svm_predict([0] * len(X1), X1, model)
delta = (datetime.now() - tstart)
# stop time
svm_results.append(delta.seconds + delta.microseconds / mu_second)
def getSVMValidationPredictions(input, output, folds): n = len(input) predictions = [None] * n for fold in range(folds): testIn = [input[x] for x in range(n) if x % folds == fold] testOut = [output[x] for x in range(n) if x % folds == fold] trainIn = [input[x] for x in range(n) if x % folds != fold] trainOut = [output[x] for x in range(n) if x % folds != fold] testIndeces = range(fold, n, folds) #print trainIn[0], trainOut[0] prob = svmutil.svm_problem(trainOut, trainIn) param = getSvmParam() model = svmutil.svm_train(prob, param) labels, acc, vals = svmutil.svm_predict([0] * len(testOut), testIn, model) for index, label in zip(testIndeces, labels): predictions[index] = label return predictions
def svm_learning_curve(x, y):
m = len(y)
n = len(x)
steep = m / 100;
training_examples = []
train_accuracy = []
validation_accuracy = []
for i in range(steep, m, steep):
prob = svmutil.svm_problem(y[:i], x[:i])
param = svmutil.svm_parameter('-t 2 -q -c 0.01')
m = svmutil.svm_train(prob, param)
p_label_train, p_acc_train, p_val_train = svmutil.svm_predict(y[:i], x[:i], m)
p_label_validation, p_acc_validation, p_val_validation = svmutil.svm_predict(y[i:], x[i:], m)
print p_acc_train[0], "\t", p_acc_validation[0], "\n"
training_examples.append(i)
train_accuracy.append(p_acc_train[0])
validation_accuracy.append(p_acc_validation[0])
return training_examples, train_accuracy, validation_accuracy
def create_svm_problem(feats, labelledRects):
labels = []
values = []
height, width, nd = feats.shape
for rect in labelledRects:
x0, y0, x1, y1 = rect.box
if x0 >= width or y0 >= height:
continue
if x1 >= width:
x1 = width - 1
if y1 >= height:
y1 = height - 1
for x in range(x0, x1 + 1):
for y in range(y0, y1 + 1):
vs = []
for i in feats[y][x]:
vs.append(i)
labels.append(rect.label)
values.append(vs)
return svmutil.svm_problem(labels, values)
def TrainSvmPoly2(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 1 -d 2 -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 1 -d 2 -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 compute_auc_kern(data, labels, k=10, C=1.0, kern=6, gamma=None):
''' This is an interface to the extended libsvm implementation
with new kernels '''
kv = cross_validation.StratifiedKFold(labels, n_folds=k, random_state=1)
s = 0.0
for train_index, test_index in kv:
data_train = data[train_index]
labels_train = labels[train_index]
data_test = data[test_index]
labels_test = labels[test_index]
y_train = labels_train.tolist()
x_train = data_train.tolist()
prob = svmutil.svm_problem(y_train, x_train)
if gamma != None:
param = svmutil.svm_parameter("-t %i -c %.410f -q -g %i" % (kern, C, gamma))
else:
param = svmutil.svm_parameter("-t %i -c %.410f -q" % (kern, C))
model = svmutil.svm_train(prob, param)
y_test = labels_test.tolist()
x_test = data_test.tolist()
p_label, p_acc, p_val = svmutil.svm_predict(y_test, x_test, m=model)
fpr, tpr, thresholds = roc_curve(labels_test, p_val, pos_label=-1.0)
AUC = roc_auc_score(labels_test, p_val)
s += AUC
return s / k
##### RBF level 1 grid #####
# param = SVM.svm_parameter('-t 2 -c 1.126 -g 11.26 -h 0') # 75%
# param = SVM.svm_parameter('-t 2 -c 112.6 -g 1.126 -h 0') # 87.5%
# param = SVM.svm_parameter('-t 2 -c 11.26 -g 1.126 -h 0') # 87.9%
# param = SVM.svm_parameter('-t 2 -c 1.126 -g 1.126 -h 0') # 86%
# param = SVM.svm_parameter('-t 2 -c 0.1126 -g 1.126 -h 0') # 79%
# param = SVM.svm_parameter('-t 2 -c 0.01126 -g 1.126 -h 0') # %
# param = SVM.svm_parameter('-t 2 -c 112.6 -g 0.1126 -h 0') # 83.9%
# param = SVM.svm_parameter('-t 2 -c 11.26 -g 0.1126 -h 0') # 83.9
# param = SVM.svm_parameter('-t 2 -c 1.126 -g 0.1126 -h 0') # 82%
# param = SVM.svm_parameter('-t 2 -c 0.1126 -g 0.1126 -h 0') # 64%
# param = SVM.svm_parameter('-t 2 -c 0.01126 -g 0.1126 -h 0') #
##### RBF level 2 grid #####
# param = SVM.svm_parameter('-t 2 -c 2 -g 1 -h 0') # 87.5%
# param = SVM.svm_parameter('-t 2 -c 5 -g 1 -h 0') # 88.06%
# param = SVM.svm_parameter('-t 2 -c 10 -g 1 -h 0') # 87.9
##### RBF level 3 grid #####
# param = SVM.svm_parameter('-t 2 -c 5 -g 0.2 -h 0') #
# param = SVM.svm_parameter('-t 2 -c 5 -g 0.5 -h 0') # 86%
# param = SVM.svm_parameter('-t 2 -c 5 -g 1 -h 0') # 88.06%
param = SVM.svm_parameter('-t 2 -c 5 -g 2 -h 0') # 88.06%
# param = SVM.svm_parameter('-t 2 -c 5 -g 5 -h 0') # 86.55%
problem = SVM.svm_problem(labTrn.tolist(), datTrn.tolist())
model = SVM.svm_train(problem, param)
SVM.svm_predict(labVal.tolist(), datVal.tolist(), model)
p_label_train, p_acc_train, p_val_train = svmutil.svm_predict(y[:i], x[:i], m)
p_label_validation, p_acc_validation, p_val_validation = svmutil.svm_predict(y[i:], x[i:], m)
print p_acc_train[0], "\t", p_acc_validation[0], "\n"
training_examples.append(i)
train_accuracy.append(p_acc_train[0])
validation_accuracy.append(p_acc_validation[0])
return training_examples, train_accuracy, validation_accuracy
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]
if __name__ == '__main__':
y, x = svmutil.svm_read_problem("char_recon_shuffled.db")
gamma = 1.0 / (2.0 * (3.0 ** 7) ** 2)
C = 3.0 ** 3.0
prob = svmutil.svm_problem(y, x)
param = svmutil.svm_parameter('-t 2 -q -c {0} -g {1}'.format(C, gamma))
m = svmutil.svm_train(prob, param)
svmutil.svm_save_model("model", m)
def _complete_training(self, debug=False):
""" Iterate over the complete data to get the initial model """
########## read complexities file if given ##########
if self.complexities_path is not None:
import yaml
complexities_file=open(self.complexities_path, 'r')
complexities = yaml.load(complexities_file)
# nr of channels = nr of features (==dim) / features_per_channel
if not 'features_per_channel' in complexities:
complexities['features_per_channel'] = 1
self.complexity = complexities[
round(self.dim/complexities['features_per_channel'])]
self._log("Read complexity %s from file. Dimension is %s" %
(self.complexity, self.dim), level=logging.INFO)
# not compatible with regression!
# self._log("Instances of Class %s: %s, %s: %s" \
# % (self.classes[0],
# self.labels.count(self.classes.index(self.classes[0])),
# self.classes[1],
# self.labels.count(self.classes.index(self.classes[1]))))
# instead this?:
self._log("Performing training of SVM.")
########## Calculation of default gamma ##########
self.calculate_gamma()
self.num_samples = len(self.samples)
# nr_weight is the number of elements in the array weight_label and
# weight. Each weight[i] corresponds to weight_label[i], meaning that
# the penalty of class weight_label[i] is scaled by a factor of
# weight[i]. If you do not want to change penalty for any of the
# classes, just set nr_weight to 0.
########## preparation of the libsvm command ##########
# for probability output add "-b 1" to options
options = \
"-c %.42f -d %d -g %.42f -r %.42f -n %.42f -p %.42f -e %.20f -m %.42f" % \
(self.complexity, self.exponent, self.gamma,
self.offset, self.nu, self.epsilon, self.tolerance, 1000)
# use 1000MB instead of 100MB (default)
# options += " -b 1" un-comment this for probabilistic output!
if self.multinomial:
options += " -b 1"
for i,w in enumerate(self.weight):
options += " -w%d %.42f" % (i, w)
if self.kernel_type == 'LINEAR':
options += " -t 0"
elif self.kernel_type == 'POLY':
options += " -t 1"
elif self.kernel_type == 'RBF':
options += " -t 2"
elif self.kernel_type == 'SIGMOID':
options += " -t 3"
else:
self.kernel_type = 'LINEAR'
options += " -t 0"
warnings.warn("Kernel unknown! Precomputed Kernels are not " +
"yet implemented. Linear Kernel used.")
# PRECOMPUTED: kernel values in training_set_file
# (not yet implemented)
if self.svm_type == 'C-SVC':
options += " -s 0"
elif self.svm_type == 'nu-SVR':
options += " -s 1"
elif self.svm_type == 'one-class SVM':
options += " -s 2"
elif self.svm_type == 'epsilon-SVR':
options += " -s 3"
else:
options += " -s 0"
self.svm_type = 'C-SVC'
warnings.warn("SVM-type unknown. C-SVC will be used!")
if not self.debug:
options += " -q"
self._log("Libsvm is now quiet!")
old_libsvm_options = options
if self.max_iterations != 0:
options += " -i %d" % self.max_iterations
try:
param = svmutil.svm_parameter(options)
except ValueError:
param = svmutil.svm_parameter(old_libsvm_options)
self._log(
"Using max_iterations is not supported by the standard " +
"LIBSVM. Change your Python path to our customized version!",
level=logging.CRITICAL)
# transform labels with *label_function*
if self.str_label_function is not None:
self.label_function = eval(self.str_label_function)
self.labels = self.label_function(self.labels)
# build the classifier
# h = [map(float,list(data)) for data in self.samples]
problem = svmutil.svm_problem(self.labels, [
map(float, list(data)) for data in self.samples])
model = svmutil.svm_train(problem, param)
if not self.multinomial:
if (self.svm_type == 'C-SVC' or self.svm_type == 'one-class SVM') \
and self.kernel_type == 'LINEAR':
self.calculate_classification_vector(model)
if self.debug:
# This calculation is needed for further analysis
self.calculate_slack_variables(model)
print "LIBSVM Parameter:"
self.print_variables()
else:
# Slack variables are the same no matter which kernel is used
# This method is mainly used to reduce the number of samples
# being stored later on.
if self.debug:
self.calculate_slack_variables(model)
self.model = model
else:
self.model = model
# Slack variables are the same no matter which kernel is used
# This method is mainly used to reduce the number of samples
# being stored later on.
# read number of iterations needed to solve the problem
if self.max_iterations != 0:
try:
predictor_iterations = model.get_num_iterations()
self.classifier_information["~~Solver_Iterations~~"] = \
predictor_iterations
if predictor_iterations == 0 or \
predictor_iterations == numpy.Inf:
self.classifier_information["~~SVM_Converged~~"] = False
else:
self.classifier_information["~~SVM_Converged~~"] = True
except:
warnings.warn("Could not read state of the LibSVM Solver " +
"from the C-Library!")
try:
self.classifier_information["~~offset~~"] = self.b
self.classifier_information["~~w0~~"] = self.w[0]
self.classifier_information["~~w1~~"] = self.w[1]
except:
pass
self.delete_training_data()
x_train, x_test = x[:3451], x[3451:]
# print(x_train)
# print(y_train)
param = svmutil.csr_find_scale_param(x_train, lower=0)
x_train = svmutil.csr_scale(x_train, param)
x_test = svmutil.csr_scale(x_test, param)
best_err_cross = 1.
best_params_cross = None
# best_acc_predict = 0
# best_params_predict = None
prob = svmutil.svm_problem(y_train, x_train)
K = 8
error_values = {}
with open("results\\10_fold_valid.txt", "w") as out_res:
for d in [1, 2, 3, 4]:
# for d in [1, 2]:
error_values[d] = {}
# for k in range(4, K+1):
for k in range(-K, K + 1):
print(f"K = {k}, d = {d}")
out_res.write(f"K = {k}, d = {d}")
params = f'-t 1 -d {d} -c {2**k} -r 1 -q'
tsx = train_set_x.get_value()
svm_inputs = [list(tsx[s, :]) for s in range(tsx.shape[0])]
# use targets to train one svm for each hidden neuron
print "Training SVMs..."
probs = []
params = []
svms = []
ws = []
bs = []
werrs = 0
for n in range(hl.n_out):
print "Hidden neuron: %d" % n,
print " Problem...",
if n == 0:
probs.append(svmutil.svm_problem(svm_targets[n], svm_inputs))
else:
probs.append(svmutil.svm_problem(svm_targets[n], None, tmpl=probs[0]))
params.append(svmutil.svm_parameter("-q -s 0 -t 0 -c 100"))
print " Training...",
svms.append(svmutil.svm_train(probs[n], params[n]))
print " Saving...",
svmutil.svm_save_model("hidden%04d.svm" % n, svms[n])
print " Testing..."
# get weights from SVM
w, b = get_svm_weights(svms[n], hl.n_in)
ws.append(w)
bs.append(b)
# test model
