6af70cbc677f0e49024d58a6d12dad19dea8ff9b,codes/3-neural_networks/recurrent-neural-networks/code/rnn.py,,,#,22

Before Change

// Optimization Flags //
////////////////////////////////////////////

parser.add_argument("--learning_rate", default=0.001, type=float, help="initial learning rate")
parser.add_argument("--seed", default=111, type=int, help="seed")

////////////////////////////////////
// Training Flags //
////////////////////////////////////
parser.add_argument("--batch_size", default=128, type=int, help="Batch size for training")
parser.add_argument("--num_epoch", default=10, type=int, help="Number of training iterations")
parser.add_argument("--batch_per_log", default=10, type=int, help="Print the log at what number of batches?")

//////////////////////////////
// Model Flags //
//////////////////////////////

After Change

////////////////////////////////////////////

tf.app.flags.DEFINE_float("learning_rate", default=0.001, help="initial learning rate")
tf.app.flags.DEFINE_integer("seed", default=111, help="seed")

////////////////////////////////////
// Training Flags //
////////////////////////////////////
tf.app.flags.DEFINE_integer("batch_size", default=128, help="Batch size for training")
tf.app.flags.DEFINE_integer("num_epoch", default=10, help="Number of training iterations")
tf.app.flags.DEFINE_integer("batch_per_log", default=10, help="Print the log at what number of batches?")

//////////////////////////////
// Model Flags //
//////////////////////////////
tf.app.flags.DEFINE_integer("hidden_size", default=128, help="Number of neurons for RNN hodden layer")

// Store all elemnts in FLAG structure!
args = tf.app.flags.FLAGS


// Reset the graph set the random numbers to be the same using "seed"
tf.reset_default_graph()

Instances