rnn_model_training Namespace Reference

Variables
	parser = argparse.ArgumentParser()
	help
	type
	required
	default
	args = parser.parse_args()
x	f = lambda x[: args.max_letters]
	df
	axis
	how
	inplace
	columns
	indexNames = df[df['state'].str.len() < args.min_letters].index
	vocabulary = list('abcdefghiABCDEFGHIrstuvwxyzRSTUVWXYZ1234567890,.+*')
dict	int_of_letters = {}
	vocabulary_size = len(int_of_letters)
int	features_per_sample = 1
	x_data = df['state'].to_numpy()
	y_data = df['label'].to_numpy()
	max_length_of_outtupple = max([len(sublist) for sublist in df.state.to_list()])
	padded_x_data
	train_x_data = padded_x_data
	train_y_data = y_data
	num_outtuples = train_x_data.shape[0]
	timesteps = max_length_of_outtupple
tuple	input_shape = (timesteps, features_per_sample)
	model = tf.keras.models.Sequential()
	loss = history.history['loss']
	optimizer
	metrics
	history
	model_file
	overwrite
	acc = history.history['accuracy']
	val_acc = history.history['val_accuracy']
	val_loss = history.history['val_loss']
	epochs = range(1, len(acc) + 1)
	label

Variable Documentation

acc

rnn_model_training.acc = history.history['accuracy']

args

rnn_model_training.args = parser.parse_args()

axis

rnn_model_training.axis

columns

rnn_model_training.columns

default

rnn_model_training.default

df

rnn_model_training.df

Initial value:

=  pd.read_csv(
        csvfile,
        delimiter='|',
        names=['note', 'label', 'model_id', 'state'],
        skipinitialspace=True,
        converters={'state': f},
    )

epochs

rnn_model_training.epochs = range(1, len(acc) + 1)

f

x rnn_model_training.f = lambda x[: args.max_letters]

features_per_sample

int rnn_model_training.features_per_sample = 1

help

rnn_model_training.help

history

rnn_model_training.history

Initial value:

=  model.fit(
    train_x_data,
    train_y_data,
    epochs=args.epochs,
    batch_size=args.batch_size,
    validation_split=0.1,
    verbose=1,
    shuffle=True,
)

how

rnn_model_training.how

indexNames

rnn_model_training.indexNames = df[df['state'].str.len() < args.min_letters].index

inplace

rnn_model_training.inplace

input_shape

tuple rnn_model_training.input_shape = (timesteps, features_per_sample)

int_of_letters

dict rnn_model_training.int_of_letters = {}

label

rnn_model_training.label

loss

rnn_model_training.loss = history.history['loss']

max_length_of_outtupple

rnn_model_training.max_length_of_outtupple = max([len(sublist) for sublist in df.state.to_list()])

metrics

rnn_model_training.metrics

model

rnn_model_training.model = tf.keras.models.Sequential()

model_file

rnn_model_training.model_file

num_outtuples

rnn_model_training.num_outtuples = train_x_data.shape[0]

optimizer

rnn_model_training.optimizer

overwrite

rnn_model_training.overwrite

padded_x_data

rnn_model_training.padded_x_data

Initial value:

=  pad_sequences(
    x_data, maxlen=max_length_of_outtupple, padding='post'
)

parser

rnn_model_training.parser = argparse.ArgumentParser()

required

rnn_model_training.required

timesteps

rnn_model_training.timesteps = max_length_of_outtupple

train_x_data

rnn_model_training.train_x_data = padded_x_data

train_y_data

rnn_model_training.train_y_data = y_data

type

rnn_model_training.type

val_acc

rnn_model_training.val_acc = history.history['val_accuracy']

val_loss

rnn_model_training.val_loss = history.history['val_loss']

vocabulary

rnn_model_training.vocabulary = list('abcdefghiABCDEFGHIrstuvwxyzRSTUVWXYZ1234567890,.+*')

vocabulary_size

rnn_model_training.vocabulary_size = len(int_of_letters)

x_data

rnn_model_training.x_data = df['state'].to_numpy()

y_data

rnn_model_training.y_data = df['label'].to_numpy()