Last active
January 12, 2017 02:27
-
-
Save tu-lee/0af2fd9d8c8d56e2c612e5ea0c3a4639 to your computer and use it in GitHub Desktop.
RNN experiement using custom exponential signal data
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import tensorflow as tf | |
import numpy as np | |
import random | |
import math | |
from mlxtend.preprocessing import one_hot | |
from tensorflow.python.ops import rnn, rnn_cell | |
""" | |
Hyperparameters | |
""" | |
hm_epochs = 10 | |
n_classes = 2 | |
batch_size = 10 | |
chunk_size = 1 | |
n_chunks = 1 | |
rnn_size = 10 | |
""" | |
Parameters for generating input multiple exponential signals for input | |
""" | |
lorange= 1 | |
resolution= 500 | |
hirange= 1000 | |
amplitude= np.random.uniform(-10,10) | |
t = 100 | |
no_tau = 100 | |
"""Input signals""" | |
for X in range(no_tau): | |
random.seed() | |
tau = np.array([int(math.ceil(np.random.uniform(lorange, hirange)))]) | |
X= np.array(amplitude * np.exp(-t / tau)) | |
X = np.reshape(X, [-1, 1,1]) | |
print(X.shape) | |
"""Output labels""" | |
count = 0 | |
while (count <no_tau): | |
tau = np.array([int(math.ceil(np.random.uniform(lorange, hirange)))]) | |
label = np.array(one_hot([int(math.ceil(tau / resolution))])) | |
count = count + 1 | |
print ('For tau value of', tau, 'label is', label) | |
""" Input placeholders for signal and label""" | |
x = tf.placeholder('float', [None,n_chunks, chunk_size]) | |
y = tf.placeholder('float') | |
"""Define RNN function""" | |
def recurrent_neural_network(x): | |
layer = {'weights':tf.Variable(tf.random_normal([rnn_size,n_classes])), | |
'biases':tf.Variable(tf.random_normal([n_classes]))} | |
x = tf.transpose(x, [1, 0, 2]) | |
x = tf.reshape(x, [-1, chunk_size]) | |
x = tf.split(0, n_chunks, x) | |
lstm_cell = rnn_cell.BasicLSTMCell(rnn_size,state_is_tuple=True) | |
outputs, states = rnn.rnn(lstm_cell, x, dtype=tf.float32) | |
outputs = tf.matmul(outputs[-1],layer['weights']) + layer['biases'] | |
return outputs | |
"""Training the network""" | |
def train_neural_network(x): | |
prediction = recurrent_neural_network(x) | |
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(prediction, y)) | |
optimizer = tf.train.GradientDescentOptimizer(0.5).minimize(cost) | |
with tf.Session() as sess: | |
sess.run(tf.global_variables_initializer()) | |
for epoch in range(hm_epochs): | |
epoch_loss = 0 | |
i = 0 | |
while i < no_tau: | |
start = i | |
end = i + batch_size | |
batch_x = np.array(X[start:end]) | |
batch_y = np.array(label[start:end]) | |
_, c = sess.run([optimizer, cost], feed_dict={x: batch_x, y: batch_y}) | |
epoch_loss += c | |
i += batch_size | |
print('Epoch', epoch + 1, 'completed out of', hm_epochs, 'loss:', epoch_loss) | |
correct = tf.equal(tf.argmax(prediction, 1), tf.argmax(y, 1)) | |
accuracy = -tf.reduce_mean(tf.cast(correct, 'float')) | |
print('Accuracy:', accuracy.eval({x: batch_x, y: batch_y})) | |
train_neural_network(x) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment