def normlize_data(self):
### normalize train data ###
if os.path.isfile(self.inp_stats_file) and os.path.isfile(
self.out_stats_file):
self.inp_scaler = data_utils.load_norm_stats(self.inp_stats_file,
self.inp_dim,
method=self.inp_norm)
self.out_scaler = data_utils.load_norm_stats(self.out_stats_file,
self.out_dim,
method=self.out_norm)
else:
print(
'preparing train_x, train_y from input and output feature files...'
)
train_x, train_y, train_flen = data_utils.read_data_from_file_list(self.inp_train_file_list, self.out_train_file_list,\
self.inp_dim, self.out_dim, sequential_training=True if self.sequential_training or self.encoder_decoder else False)
print('computing norm stats for train_x...')
inp_scaler = data_utils.compute_norm_stats(train_x,
self.inp_stats_file,
method=self.inp_norm)
print('computing norm stats for train_y...')
out_scaler = data_utils.compute_norm_stats(train_y,
self.out_stats_file,
method=self.out_norm)
def train_tensorflow_model(self):
print('preparing train_x, train_y from input and output feature files...')
#### load the data ####
train_x, train_y, train_flen = data_utils.read_data_from_file_list(self.inp_train_file_list, self.out_train_file_list,
self.inp_dim, self.out_dim, sequential_training=True if self.sequential_training or self.encoder_decoder else False)
#### normalize the data ####
data_utils.norm_data(train_x, self.inp_scaler, sequential_training=True if self.sequential_training or self.encoder_decoder else False)
data_utils.norm_data(train_y, self.out_scaler, sequential_training=True if self.sequential_training or self.encoder_decoder else False)
#### define the model ####
if self.sequential_training:
utt_length=train_flen["utt2framenum"].values()
self.tensorflow_models.get_max_step(max(utt_length))
self.tensorflow_models.define_sequence_model()
elif self.encoder_decoder:
utt_length=train_flen["utt2framenum"].values()
super(Train_Encoder_Decoder_Models,self.encoder_decoder_models).__setattr__("max_step",max(utt_length))
self.encoder_decoder_models.define_encoder_decoder()
else:
self.tensorflow_models.define_feedforward_model()
#### train the model ####
print('training...')
if self.sequential_training:
### Train feedforward model ###
self.tensorflow_models.train_sequence_model(train_x, train_y, batch_size=self.batch_size, num_of_epochs=self.num_of_epochs, shuffle_data=self.shuffle_data,utt_length=utt_length)
elif self.encoder_decoder:
self.encoder_decoder_models.train_encoder_decoder_model(train_x,train_y,batch_size=self.batch_size,num_of_epochs=self.num_of_epochs,shuffle_data=True,utt_length=utt_length)
else:
self.tensorflow_models.train_feedforward_model(train_x, train_y, batch_size=self.batch_size, num_of_epochs=self.num_of_epochs, shuffle_data=self.shuffle_data)
def train_tensorflow_model(self):
print('preparing train_x, train_y from input and output feature files...')
#### load the data ####
train_x, train_y, train_flen = data_utils.read_data_from_file_list(self.inp_train_file_list, self.out_train_file_list,
self.inp_dim, self.out_dim, sequential_training=True if self.sequential_training or self.encoder_decoder else False)
#### normalize the data ####
data_utils.norm_data(train_x, self.inp_scaler, sequential_training=True if self.sequential_training or self.encoder_decoder else False)
data_utils.norm_data(train_y, self.out_scaler, sequential_training=True if self.sequential_training or self.encoder_decoder else False)
#### define the model ####
if self.sequential_training:
utt_length=train_flen["utt2framenum"].values()
self.tensorflow_models.get_max_step(max(utt_length))
self.tensorflow_models.define_sequence_model()
elif self.encoder_decoder:
utt_length=train_flen["utt2framenum"].values()
super(Train_Encoder_Decoder_Models,self.encoder_decoder_models).__setattr__("max_step",max(utt_length))
self.encoder_decoder_models.define_encoder_decoder()
else:
self.tensorflow_models.define_feedforward_model()
#### train the model ####
print('training...')
if self.sequential_training:
### Train feedforward model ###
self.tensorflow_models.train_sequence_model(train_x, train_y, batch_size=self.batch_size, num_of_epochs=self.num_of_epochs, shuffle_data=self.shuffle_data,utt_length=utt_length)
elif self.encoder_decoder:
self.encoder_decoder_models.train_encoder_decoder_model(train_x,train_y,batch_size=self.batch_size,num_of_epochs=self.num_of_epochs,shuffle_data=True,utt_length=utt_length)
else:
self.tensorflow_models.train_feedforward_model(train_x, train_y, batch_size=self.batch_size, num_of_epochs=self.num_of_epochs, shuffle_data=self.shuffle_data)
def normlize_data(self):
### normalize train data ###
if os.path.isfile(self.inp_stats_file) and os.path.isfile(self.out_stats_file):
self.inp_scaler = data_utils.load_norm_stats(self.inp_stats_file, self.inp_dim, method=self.inp_norm)
self.out_scaler = data_utils.load_norm_stats(self.out_stats_file, self.out_dim, method=self.out_norm)
else:
print('preparing train_x, train_y from input and output feature files...')
train_x, train_y, train_flen = data_utils.read_data_from_file_list(self.inp_train_file_list, self.out_train_file_list,\
self.inp_dim, self.out_dim, sequential_training=True if self.sequential_training or self.encoder_decoder else False)
print('computing norm stats for train_x...')
inp_scaler = data_utils.compute_norm_stats(train_x, self.inp_stats_file, method=self.inp_norm)
print('computing norm stats for train_y...')
out_scaler = data_utils.compute_norm_stats(train_y, self.out_stats_file, method=self.out_norm)
