discretizer,
normalizer,
args.batch_size,
shuffle=False)
else:
# Set number of batches in one epoch
train_nbatches = 2000
val_nbatches = 1000
if args.small_part:
train_nbatches = 20
val_nbatches = 20
train_data_gen = utils.BatchGen(reader=train_reader,
discretizer=discretizer,
normalizer=normalizer,
partition=args.partition,
batch_size=args.batch_size,
steps=train_nbatches,
shuffle=True)
val_data_gen = utils.BatchGen(reader=val_reader,
discretizer=discretizer,
normalizer=normalizer,
partition=args.partition,
batch_size=args.batch_size,
steps=val_nbatches,
shuffle=False)
if args.mode == 'train':
# Prepare training
path = os.path.join(
args.output_dir, 'keras_states/' + model.final_name +
'.chunk{epoch}.test{val_loss}.state')
args.partition, discretizer,
normalizer, args.batch_size)
val_data_gen = utils.BatchGenDeepSupervisoin(val_data_loader,
args.partition, discretizer,
normalizer, args.batch_size)
else:
# Set number of batches in one epoch
train_nbatches = 2000
val_nbatches = 1000
if (args.small_part):
train_nbatches = 20
val_nbatches = 20
train_data_gen = utils.BatchGen(reader=train_reader,
discretizer=discretizer,
normalizer=normalizer,
partition=args.partition,
batch_size=args.batch_size,
steps=train_nbatches)
val_data_gen = utils.BatchGen(reader=val_reader,
discretizer=discretizer,
normalizer=normalizer,
partition=args.partition,
batch_size=args.batch_size,
steps=val_nbatches)
#val_data_gen.steps = val_reader.get_number_of_examples() // args.batch_size
#train_data_gen.steps = train_reader.get_number_of_examples() // args.batch_size
if args.mode == 'train':
# Prepare training
path = 'keras_states/' + model.final_name + '.chunk{epoch}.test{val_loss}.state'