def test_add_lstm_cell(self):
''' Add LSTM cell. '''
net = Network('LSTM')
net.set_input_layer(InputLayer(512, 1))
c, h = nns.add_lstm_cell(net, 'cell0', 512, net.INPUT_LAYER_KEY,
net.INPUT_LAYER_KEY, net.INPUT_LAYER_KEY)
c, h = nns.add_lstm_cell(net, 'cell1', 512, net.INPUT_LAYER_KEY, c, h)
c, h = nns.add_lstm_cell(net, 'cell2', 512, net.INPUT_LAYER_KEY, c, h)
num_weights = 0
for layer in net:
try:
num_weights += net[layer].total_filter_size()
except AttributeError:
pass
self.assertEqual(num_weights, 512 * 512 * 2 * 4 * 3)
def test_add_lstm_cell_not_in(self):
''' Add LSTM cell input not in. '''
net = Network('LSTM')
net.set_input_layer(InputLayer(512, 1))
with self.assertRaisesRegex(ValueError, 'add_lstm_cell: .*in.*'):
_ = nns.add_lstm_cell(net, 'cell0', 512, 'a', net.INPUT_LAYER_KEY,
net.INPUT_LAYER_KEY)
net = Network('LSTM')
net.set_input_layer(InputLayer(512, 1))
with self.assertRaisesRegex(ValueError, 'add_lstm_cell: .*in.*'):
_ = nns.add_lstm_cell(net, 'cell0', 512, net.INPUT_LAYER_KEY, 'a',
net.INPUT_LAYER_KEY)
net = Network('LSTM')
net.set_input_layer(InputLayer(512, 1))
with self.assertRaisesRegex(ValueError, 'add_lstm_cell: .*in.*'):
_ = nns.add_lstm_cell(net, 'cell0', 512, net.INPUT_LAYER_KEY,
net.INPUT_LAYER_KEY, 'a')
from nn_dataflow.core import Network
from nn_dataflow.core import InputLayer, EltwiseLayer
from nn_dataflow.nns import add_lstm_cell
'''
LSTM from GNMT.
Sutskever, Vinyals, Le, Google, NIPS 2014
'''
NN = Network('GNMT')
NN.set_input_layer(InputLayer(1000, 1))
NL = 4
# Word embedding is a simple lookup.
# Exclude or ignore embedding processing.
WE = NN.INPUT_LAYER_KEY
# layered LSTM.
X = WE
for l in range(NL):
cell = 'cell_l{}'.format(l)
C, H = add_lstm_cell(NN, cell, 1000, X)
X = H
# log(p), softmax.
NN.add('Wd', EltwiseLayer(1000, 1, 1), prevs=(X, ))
def test_add_lstm_cell_invalid_type(self):
''' Add LSTM cell with invalid type. '''
with self.assertRaisesRegexp(TypeError, 'add_lstm_cell: .*network.*'):
_ = nns.add_lstm_cell(InputLayer(512, 1), 'cell0', 512,
None, None, None)
WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE. See the BSD-3 License for more details.
You should have received a copy of the Modified BSD-3 License along with this
program. If not, see <https://opensource.org/licenses/BSD-3-Clause>.
"""
from nn_dataflow.core import Network
from nn_dataflow.core import InputLayer, EltwiseLayer
from nn_dataflow.nns import add_lstm_cell
'''
LSTM from Show and Tell.
Vinyals et al., Google, CVPR 2015
'''
NN = Network('ShowTell')
NN.set_input_layer(InputLayer(512, 1))
# Word embedding is a simple lookup.
# Exclude or ignore embedding processing.
WE = NN.INPUT_LAYER_KEY
# LSTM.
C, H = add_lstm_cell(NN, 'cell', 512, WE)
# log(p), softmax.
NN.add('Wd', EltwiseLayer(512, 1, 1), prevs=(H, ))
This program is distributed in the hope that it will be useful, but WITHOUT ANY
WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE. See the BSD-3 License for more details.
You should have received a copy of the Modified BSD-3 License along with this
program. If not, see <https://opensource.org/licenses/BSD-3-Clause>.
"""
from nn_dataflow.core import Network
from nn_dataflow.core import InputLayer, FCLayer
from nn_dataflow.nns import add_lstm_cell
'''
LSTM for phoneme classification.
Graves and Schmidhuber, 2005
'''
NN = Network('PHONEME')
NN.set_input_layer(InputLayer(26, 1))
# Input.
NN.add('We', FCLayer(26, 140), prevs=(NN.INPUT_LAYER_KEY, ))
# LSTM.
C, H = add_lstm_cell(NN, 'cell', 140, 'We')
# Output.
NN.add('Wd', FCLayer(140, 61), prevs=(H, ))
