def matchPers(p1, rgdP, conf, score = None):
global svmModel, mt_tmp, SVMfeatures
if not svmModel:
if 'featureSet' in conf:
SVMfeatures = getattr(importlib.import_module('featureSet'), conf['featureSet'])
svmModel = svm_load_model('conf/person_' + conf['featureSet'] + '.model')
else: #default
SVMfeatures = getattr(importlib.import_module('featureSet'), 'personDefault')
svmModel = svm_load_model('conf/personDefault.model')
nodeScore = nodeSim(p1, rgdP)
#pFam = conf['families'].find_one({ 'children': p1['_id']}) #find fam if p in 'children'
pFam = getFamilyFromChild(p1['_id'], conf['families'], conf['relations'])
#rgdFam = conf['match_families'].find_one({ 'children': rgdP['_id']})
rgdFam = getFamilyFromChild(rgdP['_id'], conf['match_families'], conf['match_relations'])
famScore = familySim(pFam, conf['persons'], rgdFam, conf['match_persons'])
cand_matchtxt = mt_tmp.matchtextPerson(rgdP, conf['match_persons'], conf['match_families'], conf['match_relations'])
matchtxt = mt_tmp.matchtextPerson(p1, conf['persons'], conf['families'], conf['relations'])
cosScore = cos(matchtxt, cand_matchtxt)
if score is None and 'featureSet' in conf: #score not used by deault
try: #Lucene FIX
#from luceneUtils import search
import traceback
candidates = search(matchtxt, p1['sex'], ant=100, config=conf) #Lucene search
score = 0.0
for (kid,sc) in candidates:
if str(kid) == str(rgdP['_id']):
score = sc
break
#except: #use cos instead ?? if problems running Java in Bottle
except Exception, e:
exc_type, exc_value, exc_traceback = sys.exc_info()
traceback.print_exception(exc_type, exc_value, exc_traceback)
def cons_train_sample_for_cla(filename,indexes,dic_path,glo_aff_path,result_save_path,model_path,LSA_path,LSA_model_path,decom_meas,delete):
dic_list = read_dic(dic_path,dtype=str)
glo_aff_list = read_list(glo_aff_path)
f= file(filename,'r')
fs = file(result_save_path,'w')
fd = file(dust_save_path,'w')
m= svm_load_model(model_path)
lsa_m = svm_load_model(LSA_model_path)
U = load_lsa_model(LSA_path,"U")
for line in f.readlines():
text = line.strip().split(tc_splitTag)
if len(text)!=line_length:
fd.write(line)
continue
text_temp=""
for i in indexes:
text_temp+=str_splitTag+text[i]
vec = cons_vec_for_cla(text_temp.strip().split(str_splitTag),dic_list,glo_aff_list)
y,x=cons_svm_problem(text[0],vec)
p_lab,p_acc,p_sc=svm_predict(y,x,m)
if decom_meas==1:
weight = cal_weight(p_sc[0][0])
#vec = [value*weight for value in vec ]
vec = [0]*len(vec)
for key in x[0].keys():
vec[int(key)-1]= weight*float(x[0][key])
vec = pre_doc_svds(vec,U)
y,x=cons_svm_problem(text[0],vec)
lsa_lab,lsa_acc,lsa_sc = svm_predict(y,x,lsa_m)
fs.write(text[0]+"\t"+str(p_sc[0][0])+"\t"+str(lsa_sc[0][0])+"\t"+text[1]+"\t"+text[2]+"\n")
else :
fs.write(text[0]+"\t"+str(p_sc[0][0])+"\t"+text[1]+"\t"+text[2]+"\n")
f.close()
fs.close()
def __setstate__(self, state):
self.svm_model_fp = state['svm_model_fp']
self.svm_label_map_fp = state['svm_label_map_fp']
self.train_params = state['train_params']
self.normalize = state['normalize']
# C libraries/pointers don't survive across processes.
if '__LOCAL__' in state:
fd, fp = tempfile.mkstemp()
try:
os.close(fd)
self.svm_label_map = state['__LOCAL_LABELS__']
# write model binary to file, then load via libSVM
with open(fp, 'wb') as model_f:
model_f.write(state['__LOCAL_MODEL__'])
self.svm_model = svmutil.svm_load_model(fp)
finally:
os.remove(fp)
else:
self.svm_model = None
self._reload_model()
def predict_all(request):
'''Predicts points in an array'''
width = float( request.POST.get("width", "None") )
height = float( request.POST.get("height", "None") )
model = svmutil.svm_load_model('libsvm.model')
# Get grid of points to query
points = []
for counterY in [ 1.0 / 15.0 * y for y in range(0, 15) ]:
for counterX in [ 1.0 / 15.0 * x for x in range(0, 15) ]:
points.append([counterX, counterY])
#for counterY in [ 1.0 / 10.0 * x for x in range(0, 10) ]:
# for counterX in [ 1.0 / 10.0 * y for y in range(0, 10) ]:
# label , acc, val = svmutil.svm_predict( [0], [[counterX, counterY]], model )
# results[i] = [counterX, counterY, label]
# i = i + 1
#results["length"] = i
# Get labels
labels, acc, val = svmutil.svm_predict([0] * len(points), points, model)
results = {}
for index, value in enumerate(points):
results[index] = { "x" : points[index][0],
"y" : points[index][1],
"label" : labels[index] }
results["length"] = len(points)
return json(results)
def predict(request):
predictX = float( request.POST.get("x", -1) )
predictY = float( request.POST.get("y", -1) )
predictLabel = int( request.POST.get("label", -1) )
if predictX == -1 or predictY == -1 or predictLabel == -1:
return django.http.HttpResponse("Missing Params")
points = models.Point2d.objects.all()
# Storing the information to be presented to SVM
labels = []
inputs = []
# For each point, store the information into arrays
#for p in points:
# labels.append( p.label )
# inputs.append([p.x, p.y])
#prob = svm.svm_problem(labels, inputs)
#param = svm.svm_parameter('-t 2 -c 100')
#model = svmutil.svm_train(prob, param)
#svmutil.svm_save_model('libsvm.model', model)
model = svmutil.svm_load_model('libsvm.model')
p_label , acc, val = svmutil.svm_predict([0], [[predictX, predictY]], model)
data = {'x': predictX, 'y': predictY, 'label': int( p_label[0] ) }
return json(data)
def __setstate__(self, state):
self.__dict__.update(state)
self.svm_model_elem = \
self.svm_model_uri and from_uri(self.svm_model_uri)
self.svm_label_map_elem = \
self.svm_label_map_uri and from_uri(self.svm_label_map_uri)
# C libraries/pointers don't survive across processes.
if '__LOCAL__' in state:
# These would have gotten copied into dict during the updated.
# The instance doesn't need to keep them around after this.
del self.__dict__['__LOCAL__']
del self.__dict__['__LOCAL_LABELS__']
del self.__dict__['__LOCAL_MODEL__']
fd, fp = tempfile.mkstemp()
try:
os.close(fd)
self.svm_label_map = state['__LOCAL_LABELS__']
# write model to file, then load via libSVM
with open(fp, 'wb') as model_f:
model_f.write(state['__LOCAL_MODEL__'])
fp_bytes = fp.encode('utf8')
self.svm_model = svmutil.svm_load_model(fp_bytes)
finally:
os.remove(fp)
else:
self.svm_model = None
self._reload_model()
def __init__(self, init_data_dir, logger=None):
# load model
self.model = svmutil.svm_load_model(
os.path.join(init_data_dir, "allmodel"))
self.logger = logger
if self.logger is not None:
self.logger.debug("BRISQUE image metric has been loaded")
def load_model(model_path):
"""Load a libsvm model from a path on disk.
This currently supports:
* C-SVC
* NU-SVC
* Epsilon-SVR
* NU-SVR
Parameters
----------
model_path: str
Path on disk where the libsvm model representation is.
Returns
-------
model: libsvm_model
A model of the libsvm format.
"""
if not (HAS_LIBSVM):
raise RuntimeError(
'libsvm not found. libsvm conversion API is disabled.')
from svmutil import svm_load_model # From libsvm
import os
if (not os.path.exists(model_path)):
raise IOError("Expected a valid file path. %s does not exist" %
model_path)
return svm_load_model(model_path)
def _get_classifier(svm_name=None):
"""
If need be, initializes, and then returns a classifier trained to
differentiate between different ions and water. Also returns of options for
gathering features.
To use the classifier, you will need to pass it to
svm.libsvm.svm_predict_probability. Ion prediction is already encapsulated by
predict_ion, so most users should just call that.
Parameters
----------
svm_name : str, optional
The SVM to use for prediction. By default, the SVM trained on heavy atoms
and calcium in the presence of anomalous data is used. See
chem_data/classifiers for a full list of SVMs available.
Returns
-------
svm.svm_model
The libsvm classifier used to predict the identities of ion sites.
dict of str, bool
Options to pass to ion_vector when collecting features about these sites.
tuple of ((tuple of numpy.array of float, numpy.array of float),
tuple of float)
The scaling parameters passed to scale_to.
numpy.array of bool
The features of the vector that were selected as important for
classification. Useful for both asserting that ion_vector is returning
something of the correct size as well as only selection features that
actually affect classification.
See Also
--------
svm.libsvm.svm_predict_probability
mmtbx.ions.svm.predict_ion
phenix_dev.ion_identification.nader_ml.ions_train_svms
"""
assert (svmutil is not None)
global _CLASSIFIER, _CLASSIFIER_OPTIONS
if not svm_name or str(svm_name) == "Auto":
svm_name = _DEFAULT_SVM_NAME
if svm_name not in _CLASSIFIER:
svm_path = os.path.join(CLASSIFIERS_PATH, "{}.model".format(svm_name))
options_path = os.path.join(CLASSIFIERS_PATH,
"{}_options.pkl".format(svm_name))
try:
_CLASSIFIER[svm_name] = svmutil.svm_load_model(svm_path)
except IOError as err:
if err.errno != errno.ENOENT:
raise err
else:
_CLASSIFIER[svm_name] = None
_CLASSIFIER_OPTIONS[svm_name] = (None, None, None)
_CLASSIFIER_OPTIONS[svm_name] = load(options_path)
vector_options, scaling, features = _CLASSIFIER_OPTIONS[svm_name]
return _CLASSIFIER[svm_name], vector_options, scaling, features
def classify(filename, classLabel=0):
str = "/Thu_Life/CS/SVM/data/trainData/Test_SVMFile/singleSVM_TestFile"
f = open(str, "wb")
t = VSM.TextToVector2(filename)
slabel = ("%d ") % classLabel
if len(t) > 0:
f.write(slabel)
for k in range(len(t)):
str1 = ("%d:%d ") % (t[k][0], t[k][1])
f.write(str1)
f.write("\r\n")
else:
print "The text can't be classified to the Four Labels!"
return "Can't be classified ! "
f.close()
y, x = svmutil.svm_read_problem(str)
model = svmutil.svm_load_model("../SVMTrainFile250.model")
label, b, c = svmutil.svm_predict(y, x, model)
print "label", label
if label[0] == 1:
print "类别：财经"
return "财经"
elif label[0] == 2:
print "类别：IT"
return "IT"
elif label[0] == 3:
print "类别:旅游"
return "旅游"
elif label[0] == 4:
print "类别：体育"
return "体育"
def __on_click_train__(self, instance):
if len(self.first_folder.selection) > 0:
if self.first_folder.selection[0].endswith('.model'):
self.svm = svmutil.svm_load_model(
self.first_folder.selection[0])
#self.remove_widget(self.first_folder_label)
self.remove_widget(self.first_folder)
self.remove_widget(self.second_folder_label)
self.remove_widget(self.second_folder)
self.remove_widget(self.start_training)
self.first_folder_label.text = "Carpeta de pruebas"
#self.add_widget(self.test_folder_label)
self.add_widget(self.test_folder)
self.add_widget(self.read_images)
return
if len(self.first_folder.selection) < 1 or len(
self.second_folder.selection) < 1:
popup = Popup(
title='Error',
content=Label(
text='Necesitas seleccionar una carpeta para iniciar.'),
size=(400, 100),
size_hint=(None, None))
popup.open()
else:
popup = Popup(
title='Entrenando',
content=Label(text='Realizando entrenamiento...'),
size=(400, 150),
size_hint=(None, None),
auto_dismiss=False,
)
popup.open()
self.__start_train__(popup)
def trainSVMAndSave(modelLoc, kernel, labels):
if os.path.exists(modelLoc):
return svm_load_model(modelLoc)
else:
model = trainSVM(kernel, labels)
svm_save_model(modelLoc, model)
return model
def __setstate__(self, state):
self.svm_model_fp = state['svm_model_fp']
self.svm_label_map_fp = state['svm_label_map_fp']
self.train_params = state['train_params']
self.normalize = state['normalize']
# C libraries/pointers don't survive across processes.
if '__LOCAL__' in state:
fd, fp = tempfile.mkstemp()
try:
os.close(fd)
self.svm_label_map = state['__LOCAL_LABELS__']
# write model binary to file, then load via libSVM
with open(fp, 'wb') as model_f:
model_f.write(state['__LOCAL_MODEL__'])
self.svm_model = svmutil.svm_load_model(fp)
finally:
os.remove(fp)
else:
self.svm_model = None
self._reload_model()
def get_fearure_weights(yprob, years=range(2004,2017), normalize=False, binary_features=False, top10_type='sum', reg=1e-3):
from svmutil import svm_load_model
d=(102,103)[binary_features]
W=np.array([]).reshape(0,d)
periods=[]
for y in years:
periods.append(get_year_str(y))
# alpha, SV = get_alpha_and_SV('/home/arya/out/model.'+periods[-1])
model = svm_load_model('/home/arya/out/model.'+periods[-1])
alpha = np.array(map(lambda x: abs(x[0]), model.get_sv_coef()))
SV = model.get_SV()
X=np.zeros((len(SV),d))
for i in range(len(SV)):
for k,v in SV[i].items():
if k>0:
X[i,k-1]=v
W=np.append(W,alpha.dot(X)[None,:],axis=0)
np.set_printoptions(linewidth='1000', precision=3, edgeitems=55, suppress=True)
W=W+reg
if normalize:
W=W/ W.sum(1)[:,None]
print ('UnNormalized','Normalized')[normalize], 'Feature Weights:'
print W
if top10_type=='sum':
sumW= W.sum(0)
indices = range(len(sumW))
indices.sort(lambda x,y: -cmp(sumW[x], sumW[y]))
top10=indices[:10]
elif top10_type =='genbank':
yprob = yprob / yprob.sum(0)
err0 = abs(W[:11,:]-yprob[:,0][:,None]).sum(0)
indices = range(len(err0))
indices.sort(lambda x,y: -cmp(err0[x], err0[y]))
top10=indices[:10]
elif top10_type=='geo':
yprob = yprob / yprob.sum(0)
err1 = abs(W[:11,:]-yprob[:,1][:,None]).sum(0)
indices = range(len(err1))
indices.sort(lambda x,y: -cmp(err1[x], err1[y]))
top10=indices[:10]
elif top10_type == 'all':
top10 = range(W.shape[1])
else:
print top10_type , 'not found'
exit(1)
top10=sorted(top10)
print top10, top10_type
# exit(1)
info="""
W: t x d matrix of weights which each line contains A weight correponding to time t (periods[t]
periods: t x 1 string list which each element contains the period, e.g. 2004-2008
top10: Top 10 features which has A larger sum over all the periods
"""
Data={'W':W, 'periods':periods, 'top10':top10, 'info':info}
save_data_pkl(Data, '/home/arya/out/trends{}{}.pkl'.format(('_unnormalized','_normalized')[normalize],('_integer','_binary')[binary_features]))
return W,periods, top10
def libSVM_converter(args):
import svmutil
source_model = svmutil.svm_load_model(args.source)
from onnxmltools.convert.common.data_types import FloatTensorType
onnx_model = winmltools.convert_libsvm(source_model, get_opset(args.ONNXVersion),
initial_types=[('input', FloatTensorType([1, 'None']))])
return onnx_model
def _get_classifier(svm_name=None):
"""
If need be, initializes, and then returns a classifier trained to
differentiate between different ions and water. Also returns of options for
gathering features.
To use the classifier, you will need to pass it to
svm.libsvm.svm_predict_probability. Ion prediction is already encapsulated by
predict_ion, so most users should just call that.
Parameters
----------
svm_name : str, optional
The SVM to use for prediction. By default, the SVM trained on heavy atoms
and calcium in the presence of anomalous data is used. See
chem_data/classifiers for a full list of SVMs available.
Returns
-------
svm.svm_model
The libsvm classifier used to predict the identities of ion sites.
dict of str, bool
Options to pass to ion_vector when collecting features about these sites.
tuple of ((tuple of numpy.array of float, numpy.array of float),
tuple of float)
The scaling parameters passed to scale_to.
numpy.array of bool
The features of the vector that were selected as important for
classification. Useful for both asserting that ion_vector is returning
something of the correct size as well as only selection features that
actually affect classification.
See Also
--------
svm.libsvm.svm_predict_probability
mmtbx.ions.svm.predict_ion
phenix_dev.ion_identification.nader_ml.ions_train_svms
"""
assert (svmutil is not None)
global _CLASSIFIER, _CLASSIFIER_OPTIONS
if not svm_name or str(svm_name) == "Auto" :
svm_name = _DEFAULT_SVM_NAME
if svm_name not in _CLASSIFIER:
svm_path = os.path.join(CLASSIFIERS_PATH, "{}.model".format(svm_name))
options_path = os.path.join(CLASSIFIERS_PATH,
"{}_options.pkl".format(svm_name))
try:
_CLASSIFIER[svm_name] = svmutil.svm_load_model(svm_path)
except IOError as err:
if err.errno != errno.ENOENT:
raise err
else:
_CLASSIFIER[svm_name] = None
_CLASSIFIER_OPTIONS[svm_name] = (None, None, None)
_CLASSIFIER_OPTIONS[svm_name] = load(options_path)
vector_options, scaling, features = _CLASSIFIER_OPTIONS[svm_name]
return _CLASSIFIER[svm_name], vector_options, scaling, features
def predict(request):
predictX = float(request.POST.get("x", -1))
predictY = float(request.POST.get("y", -1))
predictLabel = int(request.POST.get("label", -1))
if predictX == -1 or predictY == -1 or predictLabel == -1:
return django.http.HttpResponse("Missing Params")
points = models.Point2d.objects.all()
# Storing the information to be presented to SVM
labels = []
inputs = []
# For each point, store the information into arrays
#for p in points:
# labels.append( p.label )
# inputs.append([p.x, p.y])
#prob = svm.svm_problem(labels, inputs)
#param = svm.svm_parameter('-t 2 -c 100')
#model = svmutil.svm_train(prob, param)
#svmutil.svm_save_model('libsvm.model', model)
model = svmutil.svm_load_model('libsvm.model')
p_label, acc, val = svmutil.svm_predict([0], [[predictX, predictY]], model)
data = {'x': predictX, 'y': predictY, 'label': int(p_label[0])}
return json(data)
def predict_all(request):
'''Predicts points in an array'''
width = float(request.POST.get("width", "None"))
height = float(request.POST.get("height", "None"))
model = svmutil.svm_load_model('libsvm.model')
# Get grid of points to query
points = []
for counterY in [1.0 / 15.0 * y for y in range(0, 15)]:
for counterX in [1.0 / 15.0 * x for x in range(0, 15)]:
points.append([counterX, counterY])
#for counterY in [ 1.0 / 10.0 * x for x in range(0, 10) ]:
# for counterX in [ 1.0 / 10.0 * y for y in range(0, 10) ]:
# label , acc, val = svmutil.svm_predict( [0], [[counterX, counterY]], model )
# results[i] = [counterX, counterY, label]
# i = i + 1
#results["length"] = i
# Get labels
labels, acc, val = svmutil.svm_predict([0] * len(points), points, model)
results = {}
for index, value in enumerate(points):
results[index] = {
"x": points[index][0],
"y": points[index][1],
"label": labels[index]
}
results["length"] = len(points)
return json(results)
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 cracker(imageRaw, tempFolder, modelFilePath, mode="default"):
'''
imageRaw is the object returned by PIL.Image.open()
tempFolder is the temporary folder to store some temp files
modelFilePath is the path of trained model
if mode="quiet",then this function produces no output
This function returns the result of cracking
'''
imageGrey = imageRaw.convert("L")
imageBin = imageGrey.point(lambda x: x > 140)
CollectAndPreprocess.ridNoise(imageBin)
imagePaths = CollectAndPreprocess.cutAndSaveImage(imageBin, i, tempFolder)
result = ""
for eachPath in imagePaths:
featureList = GetFeatureAndTrain.getFeature(eachPath)
os.remove(eachPath)
#print(featureList)
featureDict = {}
for (index, value) in enumerate(featureList):
featureDict[index + 1] = value
xt = []
xt.append(featureDict)
yt = [0]
model = svmutil.svm_load_model(modelFilePath)
p_label, p_acc, p_val = svmutil.svm_predict(yt, xt, model, "-q")
for item in p_label:
result += chr(int(item))
if mode == "default":
print(result)
return result
def do_predicting(classifier_name, test_x, test_y):
model_save_file = str('./models/') + classifier_name + str('.model')
if classifier_name == 'LIBSVM':
model = svm_load_model('./models/{}.model'.format(classifier_name))
p_labels, p_acc, p_vals = svm_predict(test_y,
np.array(test_x).tolist(), model)
return p_labels
classifiers = {
'NB': naive_bayes_classifier,
'KNN': knn_classifier,
'LR': logistic_regression_classifier,
'RF': random_forest_classifier,
'DT': decision_tree_classifier,
'SVM': svm_classifier,
'SVMCV': svm_cross_validation,
'GBDT': gradient_boosting_classifier,
'ADA': ada_boosting_classifier,
'MLP': mlp_classifier
}
model = jl.load(model_save_file)[classifier_name]
predict = model.predict(test_x)
#accuracy = metrics.accuracy_score(test_y, predict)
#print 'accuracy: %.2f%%' % (100 * accuracy)
return predict
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 predict(V, yy):
m = svmutil.svm_load_model("sample.model")
x = [list(map(lambda z: z * 10, list(t))) for t in V]
y = [0 if t < 0 else 1 for t in yy]
p_label, p_acc, p_val = svmutil.svm_predict(y, x, m)
print(y)
print(p_label)
print(x[10])
def predict(V, yy):
m = svmutil.svm_load_model('sample.model')
x = ([list(map(lambda z: z * 10, list(t))) for t in V])
y = [0 if t < 0 else 1 for t in yy]
p_label, p_acc, p_val = svmutil.svm_predict(y, x, m)
print(y)
print(p_label)
print(x[10])
def doRecognize():
yt, xt = svm_read_problem(testTxt)
model = svm_load_model(cdir + "model")
p_label, p_acc, p_val = svm_predict(yt, xt, model) #p_label即为识别的结果
code = ''
for item in p_label:
code = code + str(int(item))
print code
def _reload_model(self):
"""
Reload SVM model from configured file path.
"""
if self.svm_model_fp and os.path.isfile(self.svm_model_fp):
self.svm_model = svmutil.svm_load_model(self.svm_model_fp)
if self.svm_label_map_fp and os.path.isfile(self.svm_label_map_fp):
with open(self.svm_label_map_fp, "rb") as f:
self.svm_label_map = cPickle.load(f)
def _reload_model(self):
"""
Reload SVM model from configured file path.
"""
if self.svm_model_fp and os.path.isfile(self.svm_model_fp):
self.svm_model = svmutil.svm_load_model(self.svm_model_fp)
if self.svm_label_map_fp and os.path.isfile(self.svm_label_map_fp):
with open(self.svm_label_map_fp, 'rb') as f:
self.svm_label_map = cPickle.load(f)
def think(self,text):
from twss.twss import twss_lite
import pickle
from svmutil import svm_load_model
input = open(self.vocab)
vocabList = pickle.load(input)
input.close()
model = svm_load_model(self.model)
return "That's what she said!" if twss_lite(text,vocabList,model) == 1 else ""
def get_BRISQUE_score(cv_image):
# convert to gray scale
dis = cv2.cvtColor(cv_image, cv2.COLOR_BGR2GRAY)
# compute feature vectors of the image
features = compute_features(dis)
# rescale the brisqueFeatures vector from -1 to 1
x = [0]
# pre loaded lists from C++ Module to rescale brisquefeatures vector to [-1, 1]
min_ = [
0.336999, 0.019667, 0.230000, -0.125959, 0.000167, 0.000616, 0.231000,
-0.125873, 0.000165, 0.000600, 0.241000, -0.128814, 0.000179, 0.000386,
0.243000, -0.133080, 0.000182, 0.000421, 0.436998, 0.016929, 0.247000,
-0.200231, 0.000104, 0.000834, 0.257000, -0.200017, 0.000112, 0.000876,
0.257000, -0.155072, 0.000112, 0.000356, 0.258000, -0.154374, 0.000117,
0.000351
]
max_ = [
9.999411, 0.807472, 1.644021, 0.202917, 0.712384, 0.468672, 1.644021,
0.169548, 0.713132, 0.467896, 1.553016, 0.101368, 0.687324, 0.533087,
1.554016, 0.101000, 0.689177, 0.533133, 3.639918, 0.800955, 1.096995,
0.175286, 0.755547, 0.399270, 1.095995, 0.155928, 0.751488, 0.402398,
1.041992, 0.093209, 0.623516, 0.532925, 1.042992, 0.093714, 0.621958,
0.534484
]
# append the rescaled vector to x
for i in range(0, 36):
min = min_[i]
max = max_[i]
x.append(-1 + (2.0 / (max - min) * (features[i] - min)))
# load model
model = svmutil.svm_load_model("allmodel")
# create svm node array from python list
x, idx = gen_svm_nodearray(
x[1:], isKernel=(model.param.kernel_type == PRECOMPUTED))
x[36].index = -1 # set last index to -1 to indicate the end.
# get important parameters from model
svm_type = model.get_svm_type()
is_prob_model = model.is_probability_model()
nr_class = model.get_nr_class()
if svm_type in (ONE_CLASS, EPSILON_SVR, NU_SVC):
# here svm_type is EPSILON_SVR as it's regression problem
nr_classifier = 1
dec_values = (c_double * nr_classifier)()
# calculate the quality score of the image using the model and svm_node_array
qualityscore = svmutil.libsvm.svm_predict_probability(model, x, dec_values)
return qualityscore
def predict(self, document):
"""
tag the new data basde on a given model
:param document: document object
:returns: list of predictions
"""
logger.info('Start to predict')
y,x = self.fit(document)
if Svm._auto_config:
file_name = self._auto_load('file', self.model_file)
m = svmutil.svm_load_model(file_name)
pkg_resources.cleanup_resources()
else :
m = svmutil.svm_load_model(self.model_file)
y_pred, p_acc, p_val = svmutil.svm_predict(y, x, m, "-q")
return y_pred
def think(self, text):
from twss.twss import twss_lite
import pickle
from svmutil import svm_load_model
input = open(self.vocab)
vocabList = pickle.load(input)
input.close()
model = svm_load_model(self.model)
return "That's what she said!" if twss_lite(text, vocabList,
model) == 1 else ""
def main():
try:
lModel = svmutil.svm_load_model(sys.argv[1])
lRanges = read_ranges(sys.argv[2])
lFile = sys.argv[3]
lBlockSize = int(sys.argv[4])
except IndexError, pExc:
print "Usage: " + sys.argv[0] + " <model-file> <range-file> "\
"<problem-file> <block-size>"
sys.exit(-1)
def _load_model(self) -> svmutil.svm_model:
"""
加载模型
:return: acc_model，gyro_model
"""
model_full_path = os.path.join(MODEL_DIR, self.model_name)
if not os.path.exists(model_full_path):
logger.info("模型{0}不存在，开始训练".format(model_full_path))
self._train()
return svmutil.svm_load_model(model_full_path)
def solve(im):
im = denoise(im)
im_list = crop(im)
tmp = str(round(time.time() * 1000))
with open(tmp, "w") as f:
for im in im_list:
f.write("0" + feature(im))
y, x = svmutil.svm_read_problem(tmp)
model = svmutil.svm_load_model(MODEL_FILE)
p_label, p_acc, p_val = svmutil.svm_predict(y, x, model)
os.remove(tmp)
return "".join([chr(round(x)) for x in p_label])
def _reload_model(self):
"""
Reload SVM model from configured file path.
"""
if self.svm_model_elem and not self.svm_model_elem.is_empty():
svm_model_tmp_fp = self.svm_model_elem.write_temp()
self.svm_model = svmutil.svm_load_model(svm_model_tmp_fp)
self.svm_model_elem.clean_temp()
if self.svm_label_map_elem and not self.svm_label_map_elem.is_empty():
self.svm_label_map = \
cPickle.loads(self.svm_label_map_elem.get_bytes())
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 calculate_image_quality_score(brisque_features):
model = svmutil.svm_load_model('brisque_svm.txt')
scaled_brisque_features = scale_features(brisque_features)
x, idx = svmutil.gen_svm_nodearray(
scaled_brisque_features,
isKernel=(model.param.kernel_type == svmutil.PRECOMPUTED))
nr_classifier = 1
prob_estimates = (svmutil.c_double * nr_classifier)()
return svmutil.libsvm.svm_predict_probability(model, x, prob_estimates)
def GetCode_SKB_123(picpath,types='path'):
model=svmutil.svm_load_model(modelpath_SKB_123)
if types=='path':img=Image.open(picpath).convert('L').point([0]*165+[1]*(256-165),'1')
elif types=='img':img=picpath
pixel_cnt_list=GetFeature(img,'img')
tempath=os.path.join(os.getcwd(),'temp.txt')
with open(tempath,'w') as f:
f.writelines(GetFeatureStr(pixel_cnt_list,0))
y0,x0=svmutil.svm_read_problem(tempath)
os.remove(tempath)
p_label,p_acc,p_val=svmutil.svm_predict(y0,x0,model,'-q')
return chr(int(p_label[0]))
def load_model(self, model_file, check_size=True):
"""
Taken from: https://github.com/Duke-GCB/Predict-TF-Binding
Loads a svm model from a file and computes its size
:param model_file: The file name of the model to load
:return: A dictionary with keys model, file, and size
"""
model = svmutil.svm_load_model(model_file)
model_dict = {'model': model}
if check_size:
model_dict['size'] = len(model.get_SV()[0]) - 1 # sv includes a -1 term that is not present in the model file, so subtract 1
return model_dict
def _reload_model(self):
"""
Reload SVM model from configured file path.
"""
if self.svm_model_elem and not self.svm_model_elem.is_empty():
svm_model_tmp_fp = self.svm_model_elem.write_temp()
self.svm_model = svmutil.svm_load_model(svm_model_tmp_fp)
self.svm_model_elem.clean_temp()
if self.svm_label_map_elem and not self.svm_label_map_elem.is_empty():
self.svm_label_map = \
pickle.loads(self.svm_label_map_elem.get_bytes())
def predictSVMFromFiles(tp, datasetOutDir, fileIdentifier):
jobs = []
for testKernelFileLoc in getAllKernelFileLocs(datasetOutDir, fileIdentifier, False):
testKernel = loadPickle(testKernelFileLoc)
mfl = getModelFileLoc(getKernelDir(testKernelFileLoc), 'svm', fileIdentifier)
model = svm_load_model(mfl)
predictFileLoc = getPredictFileLoc(getModelDir(mfl), fileIdentifier)
if not os.path.exists(predictFileLoc):
jobs.append(tp.apply_async(predictSVMAndSave(model, testKernel, predictFileLoc)))
return jobs
def predict_main(file, model, out, now_path):
file_path = os.path.join(now_path, file)
out_path = os.path.join(now_path, out)
y_p, x_p = svm_read_problem(file_path)
model = svm_load_model(model)
p_label, p_acc, p_val = svm_predict(y_p, x_p, model)
result = 'True,Predict\n'
for i in range(len(y_p)):
result += str(y_p[i]) + ',' + str(p_label[i]) + '\n'
with open(out_path + '_predict.csv', 'w', encoding="utf-8") as f:
f.write(result)
f.close()
def load_pretrained_targetmodels():
"""
"""
#left target model
m_target_l = []
modelfile = "model/model_targetX_left"
m = svmutil.svm_load_model(modelfile)
m_target_l.append(m)
modelfile = "model/model_targetY_left"
m = svmutil.svm_load_model(modelfile)
m_target_l.append(m)
modelfile = "model/model_targetS_left"
m = svmutil.svm_load_model(modelfile)
m_target_l.append(m)
#right target model
m_target_r = []
modelfile = "model/model_targetX_right"
m = svmutil.svm_load_model(modelfile)
m_target_r.append(m)
modelfile = "model/model_targetY_right"
m = svmutil.svm_load_model(modelfile)
m_target_r.append(m)
modelfile = "model/model_targetS_right"
m = svmutil.svm_load_model(modelfile)
m_target_r.append(m)
return [m_target_l, m_target_r]
def matchPers(p1, rgdP, conf, score=None):
global svmModel, mt_tmp, SVMfeatures
if not svmModel:
if 'featureSet' in conf:
SVMfeatures = getattr(importlib.import_module('featureSet'),
conf['featureSet'])
svmModel = svm_load_model('conf/person_' + conf['featureSet'] +
'.model')
else: #default
SVMfeatures = getattr(importlib.import_module('featureSet'),
'personDefault')
svmModel = svm_load_model('conf/personDefault.model')
nodeScore = nodeSim(p1, rgdP)
#pFam = conf['families'].find_one({ 'children': p1['_id']}) #find fam if p in 'children'
pFam = getFamilyFromChild(p1['_id'], conf['families'], conf['relations'])
#rgdFam = conf['match_families'].find_one({ 'children': rgdP['_id']})
rgdFam = getFamilyFromChild(rgdP['_id'], conf['match_families'],
conf['match_relations'])
famScore = familySim(pFam, conf['persons'], rgdFam, conf['match_persons'])
cand_matchtxt = mt_tmp.matchtextPerson(rgdP, conf['match_persons'],
conf['match_families'],
conf['match_relations'])
matchtxt = mt_tmp.matchtextPerson(p1, conf['persons'], conf['families'],
conf['relations'])
cosScore = cos(matchtxt, cand_matchtxt)
if score is None and 'featureSet' in conf: #score not used by deault
try: #Lucene FIX
#from luceneUtils import search
import traceback
candidates = search(matchtxt, p1['sex'], ant=100,
config=conf) #Lucene search
score = 0.0
for (kid, sc) in candidates:
if str(kid) == str(rgdP['_id']):
score = sc
break
#except: #use cos instead ?? if problems running Java in Bottle
except Exception, e:
exc_type, exc_value, exc_traceback = sys.exc_info()
traceback.print_exception(exc_type, exc_value, exc_traceback)
def __init__(self, name, model, labels, n_features, ftype, error):
_p, _n = labels
super(self.__class__, self).__init__(name, _p, _n, None, n_features, ftype, error)
self.model = svm_load_model(model)
def clf(x):
if hasattr(x, 'tolist'):
xx = x.tolist()
else:
xx = x
p_label, _, _ = svm_predict([0], [xx], self.model, '-q')
return p_label[0]
self.clf1 = clf
def live_test():
subdir = 'data/'
img_kinds = ["happy", "anger", "neutral", "surprise"]
models = {}
# load all the models
print "Loading Models"
for img_kind in img_kinds:
print 'loading for: ' + img_kind
models[img_kind] = svmutil.svm_load_model(subdir + img_kind + '.model')
print "---------------------"
print "Loading cascade"
face_cascade = "haarcascades/haarcascade_frontalface_alt.xml"
hc = cv.Load(face_cascade)
print "---------------------"
capture = cv.CaptureFromCAM(0)
while True:
img = cv.QueryFrame(capture)
cv.ShowImage("camera",img)
key_pressed = cv.WaitKey(50)
if key_pressed == 27:
break
elif key_pressed == 32:
print '~> KEY PRESSED <~'
# do face detection
print 'detecting face'
returned = face.handel_camera_image(img, hc)
if returned == None:
print "No face || more than one face"
pass
else:
(img_o, img_face) = returned
cv.ShowImage("face",img_face)
# get features from the face
results = {}
for img_kind in img_kinds:
test_data = get_image_features(img_face, True, img_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[img_kind])
results[img_kind] = label[0][0]
print img_kind + str(results[img_kind])
sorted_results = sorted(results.iteritems(), key=operator.itemgetter(1))
print sorted_results[len(sorted_results)-1][0]
print "---------------------"
def read_model(self,filename):
""" read a model an overwrites the old one """
try:
self.model=su.svm_load_model(filename)
except IOError:
print("Error in reading file: ", filename)
return False
m=re.search("(.*)\.(svm)",filename) # read the names from npy
basename=m.group(1)
try:
self.names=np.load(basename+".npy")
except IOError:
self.names=['x1','x2','x3']
return True
def test_model(img_kind):
subdir = "data/"
model = svmutil.svm_load_model(subdir + img_kind + '.model')
print "Finished Loading Model"
total_count = 0
correct_count = 0
wrong_count = 0
the_ones = glob.glob(subdir + "f_" + img_kind + "*.jpg")
all_of_them = glob.glob(subdir + "f_*_*.jpg")
the_others = []
for x in all_of_them:
total_count += 1
if the_ones.count(x) < 1:
the_others.append(x)
for x in the_ones:
img = cv.LoadImageM(x)
cv.ShowImage("img", img)
cv.WaitKey(10)
img_features = get_image_features(img, True, img_kind)
predict_input_data = []
predict_input_data.append(img_features)
(val, val_2, val_3) = svmutil.svm_predict([1], predict_input_data, model)
if int(val[0]) == 1:
print 'correct'
correct_count += 1
else:
wrong_count += 1
for x in the_others:
img = cv.LoadImageM(x)
cv.ShowImage("img", img)
cv.WaitKey(10)
img_features = get_image_features(img, True, img_kind)
predict_input_data = []
predict_input_data.append(img_features)
(val, val_2, val_3) = svmutil.svm_predict([1], predict_input_data, model)
if int(val[0]) == -1:
correct_count += 1
else:
wrong_count += 1
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 readModels(self, Dir):
print "Reading models..."
if not os.path.isdir(Dir):
print "[E]Cannot find out models, Please check the given directory!"
return
self.models = []
for index in xrange(self.inputDim):
filename = Dir + "/model_" + str(index)
if not os.path.isfile(filename):
print "[E]Wrong models!"
self.models = []
return
model = svmutil.svm_load_model(filename)
self.models.append(model)
print "Reading models...Done."
def test_measure_BRISQUE(imgPath):
# read image from given path
dis = cv2.imread(imgPath, 1)
if(dis is None):
print("Wrong image path given")
print("Exiting...")
sys.exit(0)
# convert to gray scale
dis = cv2.cvtColor(dis, cv2.COLOR_BGR2GRAY)
# compute feature vectors of the image
features = compute_features(dis)
# rescale the brisqueFeatures vector from -1 to 1
x = [0]
# pre loaded lists from C++ Module to rescale brisquefeatures vector to [-1, 1]
min_= [0.336999 ,0.019667 ,0.230000 ,-0.125959 ,0.000167 ,0.000616 ,0.231000 ,-0.125873 ,0.000165 ,0.000600 ,0.241000 ,-0.128814 ,0.000179 ,0.000386 ,0.243000 ,-0.133080 ,0.000182 ,0.000421 ,0.436998 ,0.016929 ,0.247000 ,-0.200231 ,0.000104 ,0.000834 ,0.257000 ,-0.200017 ,0.000112 ,0.000876 ,0.257000 ,-0.155072 ,0.000112 ,0.000356 ,0.258000 ,-0.154374 ,0.000117 ,0.000351]
max_= [9.999411, 0.807472, 1.644021, 0.202917, 0.712384, 0.468672, 1.644021, 0.169548, 0.713132, 0.467896, 1.553016, 0.101368, 0.687324, 0.533087, 1.554016, 0.101000, 0.689177, 0.533133, 3.639918, 0.800955, 1.096995, 0.175286, 0.755547, 0.399270, 1.095995, 0.155928, 0.751488, 0.402398, 1.041992, 0.093209, 0.623516, 0.532925, 1.042992, 0.093714, 0.621958, 0.534484]
# append the rescaled vector to x
for i in range(0, 36):
min = min_[i]
max = max_[i]
x.append(-1 + (2.0/(max - min) * (features[i] - min)))
# load model
model = svmutil.svm_load_model("allmodel")
# create svm node array from python list
x, idx = gen_svm_nodearray(x[1:], isKernel=(model.param.kernel_type == PRECOMPUTED))
x[36].index = -1 # set last index to -1 to indicate the end.
# get important parameters from model
svm_type = model.get_svm_type()
is_prob_model = model.is_probability_model()
nr_class = model.get_nr_class()
if svm_type in (ONE_CLASS, EPSILON_SVR, NU_SVC):
# here svm_type is EPSILON_SVR as it's regression problem
nr_classifier = 1
dec_values = (c_double * nr_classifier)()
# calculate the quality score of the image using the model and svm_node_array
qualityscore = svmutil.libsvm.svm_predict_probability(model, x, dec_values)
return qualityscore
def test(test_file, model_file, fold_num, mapping=None, multilabel=None, debug=False):
'''
Returns predicted labels, prediction values and the as the the test labels
(potentially remapped).
Requires a test instance file and a corresponding model
file. Remaps the labels in the test file (optional), classifies the test
instances against the model.
'''
if multilabel:
temp_labels, instances = alt_read_problem(test_file)
temp_labels = [[mapping[l] for l in label] for label in temp_labels]
labels = []
for temp_labs in temp_labels:
if multilabel[1] in temp_labs:
labels.append(multilabel[1])
else:
assert len(set(temp_labs)) == 1, "Something appears to be wrong with the intermediate mapping. There's still more than one label present for an instance: {0}".format(temp_labs)
labels.append([l for l in temp_labs if l != multilabel[1]][0])
else:
labels, instances = svm_read_problem(test_file)
labels = reMap(labels, mapping)
# Exclude instances which have 0 as their label
labels, instances = zip(*[(label, instance) for label, instance in zip(labels, instances) if label != 0])
if debug:
with open(os.path.basename(test_file) + '.remap', 'w') as fout:
for label, instance in zip(labels, instances):
output = '{0} '.format(str(label))
for idx, val in instance.items():
output += '{0}:{1} '.format(str(idx), str(val))
output = output.strip() + '\n'
fout.write(output)
model = svm_load_model(model_file)
print '---testing'
if classifier == 'liblinear':
pred_labels, ACC, pred_values, label_order = svm_predict(labels, instances, model)
elif classifier == 'libsvm':
pred_labels, (ACC, MSC, SCC), pred_values = svm_predict(labels, instances, model, options='-b 1')
label_order = model.get_labels()
return pred_labels, pred_values, label_order, labels
def parse_compact_nonlinear_svm(file_compact_svm, flag_load_model=True):
"""
Parse configurations and/or actual models, based on
a config file written by write_compact_nonlinear_svm.
"""
print "Loading (compact) nonlinear SVM configuration:\n%s..." % file_compact_svm
model = dict()
model["file_svm_model"] = None
model["svm_model"] = None
model["target_class"] = None
model["file_SVs"] = None
model["SVs"] = None
model["str_kernel"] = None
model["func_kernel"] = None
model_keys = model.keys()
with open(file_compact_svm) as fin:
for line in fin:
strs = line.strip().split("=")
if len(strs) == 2:
key = strs[0].strip()
if key in model_keys:
model[key] = strs[1].strip()
# string to integer
model["target_class"] = int(model["target_class"])
print model
if flag_load_model:
print "... finding kernel.."
model["func_kernel"] = getattr(kernels, model["str_kernel"])
dir_compact = os.path.dirname(file_compact_svm)
print "... loading SVM model.."
model["svm_model"] = svmutil.svm_load_model(os.path.join(dir_compact, model["file_svm_model"]))
print "... loading SVs (may take some time).."
tmp = os.path.join(dir_compact, model["file_SVs"])
if not os.path.exists(tmp):
tmp += ".npy"
model["SVs"] = np.load(tmp)
return model
def main(argv=None):
inimg = sys.argv[1] #input image file name
#load in the k-means centroids of bar chart, pie chart, and scatter plot
centroids = []
centroids.extend(pickle.load(open(centroid_bar)))
centroids.extend(pickle.load(open(centroid_pie)))
centroids.extend(pickle.load(open(centroid_scatter_plot)))
centroids = numpy.array(centroids)
D, __ = centroids.shape
img = Image.open(inimg).convert('L')
#image resize
img_resized = imageResize(img, img_size)
#get the luma of the image
luma_resized = imageToLuma(img_resized)
#get all the patches of the image
luma_patches = getAllPatches(luma_resized, patch_size)
#starndardize patches
patches_standardized = map(patchStandardize, luma_patches)
#flatten all patch matrix in patches_pool list
patches_vector = arrayLstFlatten(patches_standardized)
#get the closest centroid index based fon features
idx, __ = vq(patches_vector, centroids)
idx = idx.tolist()
#based on index form the feature
feature = map(idx.count, range(0, D))
model = svm_load_model(inmodel)
label, __, __ = svm_predict([0], [feature], model, '-q')
if label[0] == 1:
print '***bar chart***'
elif label[0] == 2:
print '***pie chart***'
elif label[0] == 3:
print '***scatter plot***'
def main():
model = sv.svm_load_model("model")
size = width, height = 400, 400
black = 0,0,0
white = 255,255,255
screen = pygame.display.set_mode(size)
pygame.display.set_caption("HW_Recon | Ingrese un caracter.")
data_flag = False
last_pos = []
while True:
for event in pygame.event.get():
if event.type == pygame.QUIT: sys.exit()
mg = pygame.mouse.get_pressed()
mpos = pygame.mouse.get_pos()
if mg == (0,0,1) and data_flag:
a = predecir(screen, model)
if DEBUG: print a
pygame.display.set_caption("HW_Recon| Letra: {0}".format(a))
if mg == (1,0,0) and not last_pos:
last_pos = mpos
data_flag = True
if mg == (1,0,0) and last_pos:
if np.linalg.norm(np.array(mpos) - np.array(last_pos)) < 30:
pygame.draw.line(screen, white, last_pos, mpos, 30)
last_pos = mpos
if mg == (0,1,0) and data_flag:
data_flag = False
last_pos = []
screen.fill(black)
pygame.display.set_caption("HW_Recon | Ingrese un caracter.")
pygame.display.flip()
def __init__(self, c=None, gamma=None, filename=None, neighbours=3, verbose=0):
self.neighbours = neighbours
self.verbose = verbose
if filename:
# If a filename is given, load a model from the given filename
if verbose:
print 'Loading classifier from "%s"...' % filename
self.model = svm_load_model(filename)
elif c == None or gamma == None:
raise Exception("Please specify both C and gamma.")
else:
self.param = svm_parameter()
self.param.C = c # Soft margin
self.param.kernel_type = RBF # Radial kernel type
self.param.gamma = gamma # Parameter for radial kernel
self.model = None
def classify2(filename, classLabel=0):
str = "/Thu_Life/CS/SVM/data/trainData/Test_SVMFile/singleSVM_TestFile"
f = open(str, "wb")
t = VSM.TextToVector2(filename)
slabel = ("%d ") % classLabel
if len(t) > 0:
f.write(slabel)
for k in range(len(t)):
str1 = ("%d:%d ") % (t[k][0], t[k][1])
f.write(str1)
f.write("\r\n")
else:
return 0
f.close()
y, x = svmutil.svm_read_problem(str)
model = svmutil.svm_load_model("../SVMTrainFile250.model")
label, b, c = svmutil.svm_predict(y, x, model)
return label[0]
def easy_predict(train_name, test_name):
range_file = train_name + ".range"
model_file = train_name + ".model"
assert os.path.exists(test_name),"testing file not found"
assert os.path.exists(model_file),"model file not found"
assert os.path.exists(range_file),"range file not found"
file_name = os.path.split(test_name)[1]
scaled_test_file = file_name + ".scale"
predict_test_file = file_name + ".predict"
cmd = '{0} -r "{1}" "{2}" > "{3}"'.format(svmscale_exe, range_file, test_name, scaled_test_file)
print('Scaling testing data...')
Popen(cmd, shell = True, stdout = PIPE).communicate()
(prob_y, prob_x) = svmutil.svm_read_problem(scaled_test_file)
model = svmutil.svm_load_model(model_file)
pred_labels, (ACC, MSE, SCC), pred_values = svmutil.svm_predict(prob_y, prob_x, model, "-b 1")
return pred_values,MSE,SCC
def iqr_model_test(filepath_model, matrix_kernel_test, clipids_test, target_class=1):
"""
Apply an SVM model on test data
@param filepath_model: a full path to load the learned SVM model
@param matrix_kernel_test: n-by-m kernel maxtrix where n (row) is |SVs| & m (col) is |test data|
@type matrix_kernel_test: 2D numpy.array
@param clipids_test: list of clipids ordered
@param target_class: positive class id. Default = 1.
@return: dictionary with 'probs','clipids'
@rtype: dictionary with 'probs' (np.array), 'clipids' (int list)
"""
model = svmutil.svm_load_model(filepath_model)
weights = svmtools.get_SV_weights_nonlinear_svm(model, target_class=target_class)
# compute margins
margins = weights[0] * matrix_kernel_test[0]
for i in range(1, matrix_kernel_test.shape[0]):
margins += weights[i] * matrix_kernel_test[i]
if matrix_kernel_test.ndim == 1: # case where there was single test data
margins = np.array([margins]) # make a single number margin into an np array
# compute probs, using platt scaling
rho = model.rho[0]
probA = model.probA[0]
probB = model.probB[0]
probs = 1.0 / (1.0 + np.exp((margins - rho) * probA + probB))
del margins
# case when the label of positive data was 2nd in SVM model symbol list
# since platt scaling was parameterized for negative data, swap probs
idx_target = svmtools.get_column_idx_for_class(model, target_class)
if idx_target == 1:
probs = 1.0 - probs
output = dict()
output['probs'] = probs
output['clipids'] = clipids_test
return output
