머신러닝스터디/2016/2016 05 14 - Revision history

imported>rabierre at 17:32, 24 July 2016

2016-07-24T17:32:41Z

← Older revision		Revision as of 17:32, 24 July 2016
Line 1:		Line 1:
	~~[[pagelist(^(~~머신러닝스터디/2016~~))]]~~		[[머신러닝스터디/2016]]
			[[머신러닝스터디/2016/목차]]

	== 내용 ==		== 내용 ==

114.108.95.133 at 05:18, 19 May 2016

2016-05-19T05:18:10Z

← Older revision		Revision as of 05:18, 19 May 2016
Line 61:		Line 61:
	print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))		print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))
	== 후기 ==		== 후기 ==
			* [[서지혜]]: 페어용 키보드는 좋은 생각이었다.
	== 다음 시간에는 ==		== 다음 시간에는 ==
	* ML Week 4 Neural Networks까지		* ML Week 4 Neural Networks까지
	== 더 보기 ==		== 더 보기 ==

114.108.95.133 at 05:16, 19 May 2016

2016-05-19T05:16:27Z

← Older revision		Revision as of 05:16, 19 May 2016
Line 62:		Line 62:
	== 후기 ==		== 후기 ==
	== 다음 시간에는 ==		== 다음 시간에는 ==
	* ML ~~동영상 보기~~		* ML Week 4 Neural Networks까지
	== 더 보기 ==		== 더 보기 ==

114.108.95.133 at 05:15, 19 May 2016

2016-05-19T05:15:11Z

← Older revision		Revision as of 05:15, 19 May 2016
Line 3:		Line 3:
	== 내용 ==		== 내용 ==
	* Tensorflow Example code에서는 Softmax를 사용했다. 우리는 이 예제를 변형해서 Sigmoid와 Cross Entropy를 직접 구현해본다.		* Tensorflow Example code에서는 Softmax를 사용했다. 우리는 이 예제를 변형해서 Sigmoid와 Cross Entropy를 직접 구현해본다.
			** [https://github.com/aymericdamien/TensorFlow-Examples/blob/master/examples/2%20-%20Basic%20Classifiers/logistic_regression.py Logistic Regression 예제코드]

	=== 코드 ===		=== 코드 ===

114.108.95.133 at 03:49, 16 May 2016

2016-05-16T03:49:37Z

← Older revision		Revision as of 03:49, 16 May 2016
Line 5:		Line 5:

	=== 코드 ===		=== 코드 ===
	# ~~python3~~		# python
	import tensorflow as tf		import tensorflow as tf
	import input_data		import input_data
Line 19:		Line 19:
	mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)		mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)
	learning_rate = 0.01		learning_rate = 0.01
	~~# x = tf.placeholder(tf.float32, shape=(None, 28, 28))~~
	x = tf.placeholder(tf.float32, shape=(None, 28 * 28))		x = tf.placeholder(tf.float32, shape=(None, 28 * 28))
	y = tf.placeholder(tf.float32, shape=(None, 10))		y = tf.placeholder(tf.float32, shape=(None, 10))

114.108.95.133 at 03:48, 16 May 2016

2016-05-16T03:48:51Z

← Older revision		Revision as of 03:48, 16 May 2016
Line 30:		Line 30:
	# h = n * 10		# h = n * 10
	h = tf.sigmoid(tf.matmul(x, weight) + bias)		h = tf.sigmoid(tf.matmul(x, weight) + bias)

	~~#cost = tf.reduce_sum(tf.add(y * tf.log(h), tf.sub(1.0, y) * tf.log(tf.sub(1.0, h)))))~~

	cost = tf.reduce_mean(-(tf.reduce_sum(y * tf.log(h), 1) + tf.reduce_sum((1. -y) * tf.log(1. -h), 1)))		cost = tf.reduce_mean(-(tf.reduce_sum(y * tf.log(h), 1) + tf.reduce_sum((1. -y) * tf.log(1. -h), 1)))

114.108.95.133 at 03:48, 16 May 2016

2016-05-16T03:48:17Z

← Older revision		Revision as of 03:48, 16 May 2016
Line 2:		Line 2:

	== 내용 ==		== 내용 ==
			* Tensorflow Example code에서는 Softmax를 사용했다. 우리는 이 예제를 변형해서 Sigmoid와 Cross Entropy를 직접 구현해본다.
	* ~~Softmax가 아니라~~ Sigmoid와 Cross Entropy를 직접 구현해본다!

	=== 코드 ===		=== 코드 ===

114.108.95.133 at 03:47, 16 May 2016

2016-05-16T03:47:39Z

← Older revision		Revision as of 03:47, 16 May 2016
Line 3:		Line 3:
	== 내용 ==		== 내용 ==

	~~== 코드 ==~~		* Softmax가 아니라 Sigmoid와 Cross Entropy를 직접 구현해본다!

			=== 코드 ===
	# python3		# python3
	import tensorflow as tf		import tensorflow as tf

114.108.95.133 at 03:46, 16 May 2016

2016-05-16T03:46:01Z

← Older revision		Revision as of 03:46, 16 May 2016
Line 62:		Line 62:
	accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))		accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))
	print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))		print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))
			== 후기 ==
			== 다음 시간에는 ==
			* ML 동영상 보기
			== 더 보기 ==

219.241.59.131: {CREATE}

2016-05-14T11:25:02Z

{CREATE}

New page

[[pagelist(^(머신러닝스터디/2016))]]

== 내용 ==

== 코드 ==

# python3
import tensorflow as tf
import input_data

if __name__ == "__main__":

# Logistic Regression
# use sigmoid

# Input
# MNIST
# Cross Entropy
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)
learning_rate = 0.01
# x = tf.placeholder(tf.float32, shape=(None, 28, 28))
x = tf.placeholder(tf.float32, shape=(None, 28 * 28))
y = tf.placeholder(tf.float32, shape=(None, 10))

# x = n * (28 * 28)
# weight = (28 * 28) * 10
weight = tf.Variable(tf.zeros((28 * 28, 10)))
bias = tf.Variable(tf.zeros((10,)))

# h = n * 10
h = tf.sigmoid(tf.matmul(x, weight) + bias)

#cost = tf.reduce_sum(tf.add(y * tf.log(h), tf.sub(1.0, y) * tf.log(tf.sub(1.0, h)))))

cost = tf.reduce_mean(-(tf.reduce_sum(y * tf.log(h), 1) + tf.reduce_sum((1. -y) * tf.log(1. -h), 1)))
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)

init = tf.initialize_all_variables()
batch_size = 100
display_step = 1
with tf.Session() as sess:
sess.run(init)

# Training cycle
for epoch in range(10):
avg_cost = 0.
total_batch = int(mnist.train.num_examples/batch_size)
# Loop over all batches
for i in range(total_batch):
batch_xs, batch_ys = mnist.train.next_batch(batch_size)
# Fit training using batch data
sess.run(optimizer, feed_dict={x: batch_xs, y: batch_ys})
# Compute average loss
avg_cost += sess.run(cost, feed_dict={x: batch_xs, y: batch_ys})/total_batch
# Display logs per epoch step
if epoch % display_step == 0:
print( "Epoch:", '%04d' % (epoch+1), "cost=", "{:.9f}".format(avg_cost))

# Test model
correct_prediction = tf.equal(tf.argmax(h, 1), tf.argmax(y, 1))
# Calculate accuracy
accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))
print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))