def eval_nn(inputp, config, output, normalization, network_type, model_file): \
# pylint: disable=too-many-arguments
""" evaluate a dataset with a neural network stored on disk
arguments:
inputp -- path to the ROOT dataset
model -- path to the yaml keras model config file
config -- path to the branches config file
output -- output name for the sqlite database (overwrites the 'test' table)
normalization -- path to the txt file with normalization constants
"""
model = keras.models.load_model(
#'Outputs/Try_3p/MDN_Three1Gauss_MC16_v3-150ep.h5', #MDN_Two1Gauss_MC16_v6-250ep.h5, 'Outputs/Try_2DGauss/pos1_MC16.h5'
#'/home/elham/NN_optimise/pixel-MDN-training/Outputs/Try_2DGauss/pos1_MC16.h5',
model_file,
custom_objects={'loss': mixture_density_loss(nb_components=1)})
_eval_dataset(model=model,
network_type=network_type,
path=inputp,
tree='NNinput',
branches=utils.get_data_config_names(config, meta=True),
norm=utils.load_normalization(normalization),
dbpath=output)
def input_fn(dataset_filename,
vocab_filename,
norm_filename=None,
num_channels=39,
batch_size=8,
num_epochs=1,
binf2phone=None,
num_parallel_calls=32,
max_frames=-1,
max_symbols=-1):
binary_targets = binf2phone is not None
labels_shape = [] if not binary_targets else len(binf2phone.index)
labels_dtype = tf.string if not binary_targets else tf.float32
dataset = utils.read_dataset(dataset_filename,
num_channels,
labels_shape=labels_shape,
labels_dtype=labels_dtype)
vocab_table = utils.create_vocab_table(vocab_filename)
if norm_filename is not None:
means, stds = utils.load_normalization(args.norm)
else:
means = stds = None
sos = binf2phone[utils.SOS].values if binary_targets else utils.SOS
eos = binf2phone[utils.EOS].values if binary_targets else utils.EOS
dataset = utils.process_dataset(dataset,
vocab_table,
sos,
eos,
means,
stds,
batch_size,
num_epochs,
binary_targets=binary_targets,
labels_shape=labels_shape,
num_parallel_calls=num_parallel_calls,
max_frames=max_frames,
max_symbols=max_symbols)
return dataset
def input_fn(dataset_filename,
vocab_filename,
norm_filename=None,
num_channels=39,
batch_size=8,
take=0,
binf2phone=None):
binary_targets = binf2phone is not None
labels_shape = [] if not binary_targets else len(binf2phone.index)
labels_dtype = tf.string if not binary_targets else tf.float32
dataset = utils.read_dataset(dataset_filename,
num_channels,
labels_shape=labels_shape,
labels_dtype=labels_dtype)
vocab_table = utils.create_vocab_table(vocab_filename)
if norm_filename is not None:
means, stds = utils.load_normalization(args.norm)
else:
means = stds = None
sos = binf2phone[utils.SOS].values if binary_targets else utils.SOS
eos = binf2phone[utils.EOS].values if binary_targets else utils.EOS
dataset = utils.process_dataset(dataset,
vocab_table,
sos,
eos,
means,
stds,
batch_size,
1,
binary_targets=binary_targets,
labels_shape=labels_shape,
is_infer=True)
if args.take > 0:
dataset = dataset.take(take)
return dataset
def input_fn(dataset_filename,
vocab_filename,
norm_filename=None,
num_channels=39,
batch_size=8,
num_epochs=1,
num_parallel_calls=32,
max_frames=-1,
max_symbols=-1,
take=0,
is_infer=False):
dataset = read_dataset(dataset_filename, num_channels)
vocab_table = utils.create_vocab_table(vocab_filename)
if norm_filename is not None and os.path.exists(norm_filename):
means, stds = utils.load_normalization(norm_filename)
else:
means = stds = None
sos = utils.SOS
eos = utils.EOS
dataset = process_dataset(dataset,
vocab_table,
sos,
eos,
means,
stds,
batch_size,
num_epochs,
num_parallel_calls=num_parallel_calls,
max_frames=max_frames,
max_symbols=max_symbols)
if take > 0:
dataset = dataset.take(take)
return dataset
def input_fn(features,
vocab_filename,
norm_filename=None,
num_channels=39,
batch_size=8,
ground_truth=None):
def gen():
if ground_truth is not None:
iterable = zip(features, ground_truth)
else:
iterable = features
for item in iterable:
yield item
output_types = (tf.float32,
tf.string) if ground_truth is not None else tf.float32
output_shapes = tf.TensorShape([None, features[0].shape[-1]])
if ground_truth is not None:
output_shapes = (output_shapes,
tf.TensorShape([None, ground_truth[0].shape[-1]]))
dataset = tf.data.Dataset.from_generator(gen, output_types, output_shapes)
vocab_table = utils.create_vocab_table(vocab_filename)
if norm_filename is not None:
means, stds = utils.load_normalization(norm_filename)
else:
means = stds = None
dataset = utils.process_dataset(dataset,
vocab_table,
utils.SOS,
utils.EOS,
means,
stds,
min(features[0].shape[0], batch_size),
1,
is_infer=True)
return dataset
def input_fn(features, vocab_filename, norm_filename=None):
def gen():
for item in features:
yield item
output_shapes = tf.TensorShape([None, features[0].shape[-1]])
dataset = tf.data.Dataset.from_generator(gen, tf.float32, output_shapes)
vocab_table = utils.create_vocab_table(vocab_filename)
if norm_filename is not None:
means, stds = utils.load_normalization(norm_filename)
else:
means = stds = None
dataset = utils.process_dataset(dataset,
vocab_table,
utils.SOS,
utils.EOS,
means,
stds,
1,
1,
is_infer=True)
return dataset