def test_model_get_outputs_rnn(backend_default, data):
data_path = load_ptb_test(path=data)
data_set = Text(time_steps=50, path=data_path)
# weight initialization
init = Constant(0.08)
# model initialization
layers = [
Recurrent(150, init, activation=Logistic()),
Affine(len(data_set.vocab), init, bias=init, activation=Rectlin())
]
model = Model(layers=layers)
output = model.get_outputs(data_set)
assert output.shape == (data_set.ndata, data_set.seq_length,
data_set.nclass)
# since the init are all constant and model is un-trained:
# along the feature dim, the values should be all the same
assert np.allclose(output[0, 0], output[0, 0, 0], rtol=0, atol=1e-5)
assert np.allclose(output[0, 1], output[0, 1, 0], rtol=0, atol=1e-5)
# along the time dim, the values should be increasing:
assert np.alltrue(output[0, 2] > output[0, 1])
assert np.alltrue(output[0, 1] > output[0, 0])
def test_model_get_outputs_rnn(backend_default, data):
data_path = load_ptb_test(path=data)
data_set = Text(time_steps=50, path=data_path)
# weight initialization
init = Constant(0.08)
# model initialization
layers = [
Recurrent(150, init, activation=Logistic()),
Affine(len(data_set.vocab), init, bias=init, activation=Rectlin())
]
model = Model(layers=layers)
output = model.get_outputs(data_set)
assert output.shape == (
data_set.ndata, data_set.seq_length, data_set.nclass)
# since the init are all constant and model is un-trained:
# along the feature dim, the values should be all the same
assert np.allclose(output[0, 0], output[0, 0, 0], rtol=0, atol=1e-5)
assert np.allclose(output[0, 1], output[0, 1, 0], rtol=0, atol=1e-5)
# along the time dim, the values should be increasing:
assert np.alltrue(output[0, 2] > output[0, 1])
assert np.alltrue(output[0, 1] > output[0, 0])
def load_ptb_test(path):
"""
Deprecated, moved to neon.data.dataloaders.
"""
logger.error('This function has moved, import from neon.data.dataloaders')
from neon.data.dataloaders import load_ptb_test # noqa
return load_ptb_test(path)
choices=['gru', 'lstm'],
help='type of recurrent layer to use (gru or lstm)')
args = parser.parse_args(gen_be=False)
# hyperparameters from the reference
args.batch_size = 20
time_steps = 20
hidden_size = 200
gradient_clip_norm = 5
# setup backend
be = gen_backend(**extract_valid_args(args, gen_backend))
# download penn treebank
train_path = load_ptb_train(path=args.data_dir)
valid_path = load_ptb_test(path=args.data_dir)
# define a custom function to parse the input into individual tokens, which for
# this data, splits into individual words. This can be passed into the Text
# object during dataset creation as seen below.
def tokenizer(s):
return s.replace('\n', '<eos>').split()
# load data and parse on word-level
train_set = Text(time_steps,
train_path,
tokenizer=tokenizer,
onehot_input=False)
valid_set = Text(time_steps,
parser.add_argument('--rlayer_type', default='lstm', choices=['gru', 'lstm'],
help='type of recurrent layer to use (gru or lstm)')
args = parser.parse_args(gen_be=False)
# hyperparameters from the reference
args.batch_size = 20
time_steps = 20
hidden_size = 200
gradient_clip_norm = 5
# setup backend
be = gen_backend(**extract_valid_args(args, gen_backend))
# download penn treebank
train_path = load_ptb_train(path=args.data_dir)
valid_path = load_ptb_test(path=args.data_dir)
# define a custom function to parse the input into individual tokens, which for
# this data, splits into individual words. This can be passed into the Text
# object during dataset creation as seen below.
def tokenizer(s):
return s.replace('\n', '<eos>').split()
# load data and parse on word-level
train_set = Text(time_steps, train_path, tokenizer=tokenizer, onehot_input=False)
valid_set = Text(time_steps, valid_path, vocab=train_set.vocab, tokenizer=tokenizer,
onehot_input=False)
# weight initialization
init = Uniform(low=-0.1, high=0.1)
def load_ptb_test(path):
logger.error('This function has moved, import from neon.data.dataloaders')
from neon.data.dataloaders import load_ptb_test # noqa
return load_ptb_test(path)
