def __init__(self, which_set, numOfClasses,
numOfExamplesPerClass, axes=('c', 0, 1, 'b')):
self.height = 32
self.width = 100
self.axes = axes
self.dtype = 'uint8'
self.examples = []
self.img_shape = (1, self.height, self.width)
self.img_size = numpy.prod(self.img_shape)
self.numOfClasses = numOfClasses
self.numOfExamplesPerClass = numOfExamplesPerClass
self.examplesPerClassCount = {}
self.which_set = which_set
if which_set == "train":
self.fileToLoadFrom = "annotation_train.txt"
elif which_set == "test":
self.fileToLoadFrom = "annotation_test.txt"
elif which_set == "valid":
self.fileToLoadFrom = "annotation_val.txt"
else:
raise ValueError("Set not recognized")
self.datapath = Config.getDatapath()
self.preprocess = Config.doPreprocess()
self.loadData()
random.seed()
view_converter = dense_design_matrix.DefaultViewConverter((self.height, self.width, 1),
axes)
super(MJSYNTH, self).__init__(X=numpy.cast['float32'](self.x), y=self.y, view_converter=view_converter,
y_labels=self.numOfClasses)
def __init__(self,
which_set,
numOfClasses,
numOfExamplesPerClass,
axes=('c', 0, 1, 'b')):
self.height = 32
self.width = 100
self.axes = axes
self.dtype = 'uint8'
self.examples = []
self.img_shape = (1, self.height, self.width)
self.img_size = numpy.prod(self.img_shape)
self.numOfClasses = numOfClasses
self.numOfExamplesPerClass = numOfExamplesPerClass
self.examplesPerClassCount = {}
self.which_set = which_set
if which_set == "train":
self.fileToLoadFrom = "annotation_train.txt"
elif which_set == "test":
self.fileToLoadFrom = "annotation_test.txt"
elif which_set == "valid":
self.fileToLoadFrom = "annotation_val.txt"
else:
raise ValueError("Set not recognized")
self.datapath = Config.getDatapath()
self.preprocess = Config.doPreprocess()
self.loadData()
random.seed()
view_converter = dense_design_matrix.DefaultViewConverter(
(self.height, self.width, 1), axes)
super(MJSYNTH, self).__init__(X=numpy.cast['float32'](self.x),
y=self.y,
view_converter=view_converter,
y_labels=self.numOfClasses)
from pylearn2.config import yaml_parse
from pylearn2.datasets.mjsynth.mjsynth import MJSYNTH
from pylearn2.datasets.mjsynth.config import Config
def output_file_string(params):
string = []
for key in params.keys():
string.append(key + ": " + str(params[key]))
return ' '.join(string)
p = Parser()
while p.has_other_configurations():
with open(Config.getYamlFilename(), 'r') as f:
yaml_file = f.read()
hyper_params = p.get_next_configuration()
outputFile = open('tests/' + p.get_num_configuration().__str__(), 'w')
sys.stdout = outputFile
sys.stderr = outputFile
yaml_file = yaml_file % (hyper_params)
print yaml_file
train = yaml_parse.load(yaml_file)
try:
train.main_loop()
except Exception as e:
from pylearn2.datasets.mjsynth.mjsynth import MJSYNTH
from pylearn2.datasets.mjsynth.config import Config
def output_file_string(params):
string = []
for key in params.keys():
string.append(key + ": " + str(params[key]))
return " ".join(string)
p = Parser()
while p.has_other_configurations():
with open(Config.getYamlFilename(), "r") as f:
yaml_file = f.read()
hyper_params = p.get_next_configuration()
outputFile = open("tests/" + p.get_num_configuration().__str__(), "w")
sys.stdout = outputFile
sys.stderr = outputFile
yaml_file = yaml_file % (hyper_params)
print yaml_file
train = yaml_parse.load(yaml_file)
try:
train.main_loop()
except Exception as e: