# deal with dropout
do_dropout = False
dropout_factor = [0.0]
input_dropout_factor = 0.0
if arguments.has_key('dropout_factor'):
do_dropout = True
if arguments.has_key('input_dropout_factor'):
input_dropout_factor = float(arguments['input_dropout_factor'])
dropout_factor = []
factors = arguments['dropout_factor'].split(',')
for n in xrange(len(factors)):
dropout_factor.append(float(factors[n]))
train_dataset, train_dataset_args = read_data_args(train_data_spec)
valid_dataset, valid_dataset_args = read_data_args(valid_data_spec)
train_sets, train_xy, train_x, train_y = read_dataset(train_dataset, train_dataset_args)
valid_sets, valid_xy, valid_x, valid_y = read_dataset(valid_dataset, valid_dataset_args)
numpy_rng = numpy.random.RandomState(89677)
theano_rng = RandomStreams(numpy_rng.randint(2 ** 30))
log('> ... building the model')
if do_dropout:
dnn = DNN_Dropout(numpy_rng=numpy_rng, theano_rng = theano_rng, n_ins=n_ins,
hidden_layers_sizes=hidden_layers_sizes, n_outs=n_outs,
activation = activation, dropout_factor = dropout_factor, input_dropout_factor = input_dropout_factor,
do_maxout = do_maxout, pool_size = pool_size,
max_col_norm = max_col_norm, l1_reg = l1_reg, l2_reg = l2_reg)
else:
dnn = DNN(numpy_rng=numpy_rng, theano_rng = theano_rng, n_ins=n_ins,
hidden_layers_sizes=hidden_layers_sizes, n_outs=n_outs,
activation = activation, do_maxout = do_maxout, pool_size = pool_size,
n_outs = int(nnet_layers[-1])
ptr_layer_number = len(hidden_layers_sizes)
if arguments.has_key('ptr_layer_number'):
ptr_layer_number = int(arguments['ptr_layer_number'])
first_layer_gb = True
if arguments.has_key('first_layer_type'):
if arguments['first_layer_type'] == 'bb':
first_layer_gb = False
# if initialized with current
keep_layer_num=0
current_nnet = wdir + 'nnet.ptr.current'
train_sets, train_xy, train_x, train_y = read_dataset(dataset, dataset_args)
# numpy random generator
numpy_rng = numpy.random.RandomState(89677)
theano_rng = RandomStreams(numpy_rng.randint(2 ** 30))
log('> ... building the model')
# construct the stacked denoising autoencoder class
srbm = SRBM(numpy_rng=numpy_rng, theano_rng = theano_rng, n_ins=n_ins,
hidden_layers_sizes=hidden_layers_sizes,
n_outs=n_outs, first_layer_gb = first_layer_gb)
if keep_layer_num > 0:
log('> ... initializing model from ' + current_nnet)
_file2nnet(srbm.sigmoid_layers, set_layer_num = keep_layer_num, filename = current_nnet, withfinal=False)
# PRETRAINING THE MODEL #
