티스토리 뷰
딥러닝으로 MNIST 98%이상 해보기¶
이번 시간은 neural net을 사용할 때 유용한 팁에 대해서 알아보겠다.¶
MNIST Softmax!¶
In [1]:
# Lab 7 Learning rate and Evaluation
import tensorflow as tf
import random
# import matplotlib.pyplot as plt
from tensorflow.examples.tutorials.mnist import input_data
tf.set_random_seed(777) # reproducibility
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)
# Check out https://www.tensorflow.org/get_started/mnist/beginners for
# more information about the mnist dataset
# parameters
learning_rate = 0.001
training_epochs = 15
batch_size = 100
# input place holders
X = tf.placeholder(tf.float32, [None, 784])
Y = tf.placeholder(tf.float32, [None, 10])
# weights & bias for nn layers
W = tf.Variable(tf.random_normal([784, 10]))
b = tf.Variable(tf.random_normal([10]))
hypothesis = tf.matmul(X, W) + b
# define cost/loss & optimizer
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(
logits=hypothesis, labels=Y))
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)
# initialize
sess = tf.Session()
sess.run(tf.global_variables_initializer())
# train my model
for epoch in range(training_epochs):
avg_cost = 0
total_batch = int(mnist.train.num_examples / batch_size)
for i in range(total_batch):
batch_xs, batch_ys = mnist.train.next_batch(batch_size)
feed_dict = {X: batch_xs, Y: batch_ys}
c, _ = sess.run([cost, optimizer], feed_dict=feed_dict)
avg_cost += c / total_batch
print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.9f}'.format(avg_cost))
print('Learning Finished!')
# Test model and check accuracy
correct_prediction = tf.equal(tf.argmax(hypothesis, 1), tf.argmax(Y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
print('Accuracy:', sess.run(accuracy, feed_dict={
X: mnist.test.images, Y: mnist.test.labels}))
# Get one and predict
r = random.randint(0, mnist.test.num_examples - 1)
print("Label: ", sess.run(tf.argmax(mnist.test.labels[r:r + 1], 1)))
print("Prediction: ", sess.run(
tf.argmax(hypothesis, 1), feed_dict={X: mnist.test.images[r:r + 1]}))
# plt.imshow(mnist.test.images[r:r + 1].
# reshape(28, 28), cmap='Greys', interpolation='nearest')
# plt.show()
지난번 MNIST data (https://jfun.tistory.com/169)에서 했던 내용이다.
hypothesis = tf.matmul(X, W) + b
# define cost/loss & optimizer
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(
logits=hypothesis, labels=Y))
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)
이 코드는 위에있는 간단한 네트워크의 모델부분인데 3줄밖에 안되는데도 정확도가 무려 90%가 나왔다.
이 모델부분을 조금 더 깊이 들어가보자.
NN for MNIST¶
3단+ReLU추가해보자.
NN에서는 레이어의 크기를 조심해야 한다.
In [2]:
# input place holders
X = tf.placeholder(tf.float32, [None, 784])
Y = tf.placeholder(tf.float32, [None, 10])
# weights & bias for nn layers
W1 = tf.Variable(tf.random_normal([784, 256]))
b1 = tf.Variable(tf.random_normal([256]))
L1 = tf.nn.relu(tf.matmul(X, W1) + b1)
W2 = tf.Variable(tf.random_normal([256, 256]))
b2 = tf.Variable(tf.random_normal([256]))
L2 = tf.nn.relu(tf.matmul(L1, W2) + b2)
W3 = tf.Variable(tf.random_normal([256, 10]))
b3 = tf.Variable(tf.random_normal([10]))
hypothesis = tf.matmul(L2, W3) + b3
# define cost/loss &/ optimizer
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(
logits=hypothesis, labels=Y))
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)
In [3]:
# initialize
sess = tf.Session()
sess.run(tf.global_variables_initializer())
# train my model
for epoch in range(training_epochs):
avg_cost = 0
total_batch = int(mnist.train.num_examples / batch_size)
for i in range(total_batch):
batch_xs, batch_ys = mnist.train.next_batch(batch_size)
feed_dict = {X: batch_xs, Y: batch_ys}
c, _ = sess.run([cost, optimizer], feed_dict=feed_dict)
avg_cost += c / total_batch
print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.9f}'.format(avg_cost))
print('Learning Finished!')
# Test model and check accuracy
correct_prediction = tf.equal(tf.argmax(hypothesis, 1), tf.argmax(Y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
print('Accuracy:', sess.run(accuracy, feed_dict={
X: mnist.test.images, Y: mnist.test.labels}))
# Get one and predict
r = random.randint(0, mnist.test.num_examples - 1)
print("Label: ", sess.run(tf.argmax(mnist.test.labels[r:r + 1], 1)))
print("Prediction: ", sess.run(
tf.argmax(hypothesis, 1), feed_dict={X: mnist.test.images[r:r + 1]}))
# plt.imshow(mnist.test.images[r:r + 1].
# reshape(28, 28), cmap='Greys', interpolation='nearest')
# plt.show()
무려 94%라니!!!!!!!!!!
지난시간에 초기화를 잘 해야 한다는 말을 한적이 있었는데, xavier라는 방법이 있었다.
모르는게 생긴다면 구글에 가서 질문하기 'xavier initialization tensorflow'
Xavier fo MNIST¶
In [4]:
# input place holders
X = tf.placeholder(tf.float32, [None, 784])
Y = tf.placeholder(tf.float32, [None, 10])
# weights & bias for nn layers
# http://stackoverflow.com/questions/33640581/how-to-do-xavier-initialization-on-tensorflow
W1 = tf.get_variable("W1", shape=[784, 256],
initializer=tf.contrib.layers.xavier_initializer())
b1 = tf.Variable(tf.random_normal([256]))
L1 = tf.nn.relu(tf.matmul(X, W1) + b1)
W2 = tf.get_variable("W2", shape=[256, 256],
initializer=tf.contrib.layers.xavier_initializer())
b2 = tf.Variable(tf.random_normal([256]))
L2 = tf.nn.relu(tf.matmul(L1, W2) + b2)
W3 = tf.get_variable("W3", shape=[256, 10],
initializer=tf.contrib.layers.xavier_initializer())
b3 = tf.Variable(tf.random_normal([10]))
hypothesis = tf.matmul(L2, W3) + b3
# define cost/loss & optimizer
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(
logits=hypothesis, labels=Y))
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)
In [5]:
# initialize
sess = tf.Session()
sess.run(tf.global_variables_initializer())
# train my model
for epoch in range(training_epochs):
avg_cost = 0
total_batch = int(mnist.train.num_examples / batch_size)
for i in range(total_batch):
batch_xs, batch_ys = mnist.train.next_batch(batch_size)
feed_dict = {X: batch_xs, Y: batch_ys}
c, _ = sess.run([cost, optimizer], feed_dict=feed_dict)
avg_cost += c / total_batch
print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.9f}'.format(avg_cost))
print('Learning Finished!')
# Test model and check accuracy
correct_prediction = tf.equal(tf.argmax(hypothesis, 1), tf.argmax(Y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
print('Accuracy:', sess.run(accuracy, feed_dict={
X: mnist.test.images, Y: mnist.test.labels}))
# Get one and predict
r = random.randint(0, mnist.test.num_examples - 1)
print("Label: ", sess.run(tf.argmax(mnist.test.labels[r:r + 1], 1)))
print("Prediction: ", sess.run(
tf.argmax(hypothesis, 1), feed_dict={X: mnist.test.images[r:r + 1]}))
# plt.imshow(mnist.test.images[r:r + 1].
# reshape(28, 28), cmap='Greys', interpolation='nearest')
# plt.show()
와 거의 98%에 육박하는 정확도가 낮았다.
그리고 재미있는점은 에폭이 1이어도 cost가 매우 낮다.
이것은 초기값이 잘 initialize 되었다는 의미이다. (같은 모델에 초기값만 잘 잡아줘도 정확도를 올리는데 영향을 많이 준다.)
Deep NN for MNIST¶
256에서 512로 넓게 하고, 그리고 5단으로 보다 깊게 해보자.
In [6]:
# input place holders
X = tf.placeholder(tf.float32, [None, 784])
Y = tf.placeholder(tf.float32, [None, 10])
# weights & bias for nn layers
# http://stackoverflow.com/questions/33640581/how-to-do-xavier-initialization-on-tensorflow
W1 = tf.get_variable("W1_", shape=[784, 512],
initializer=tf.contrib.layers.xavier_initializer())
b1 = tf.Variable(tf.random_normal([512]))
L1 = tf.nn.relu(tf.matmul(X, W1) + b1)
W2 = tf.get_variable("W2_", shape=[512, 512],
initializer=tf.contrib.layers.xavier_initializer())
b2 = tf.Variable(tf.random_normal([512]))
L2 = tf.nn.relu(tf.matmul(L1, W2) + b2)
W3 = tf.get_variable("W3_", shape=[512, 512],
initializer=tf.contrib.layers.xavier_initializer())
b3 = tf.Variable(tf.random_normal([512]))
L3 = tf.nn.relu(tf.matmul(L2, W3) + b3)
W4 = tf.get_variable("W4_", shape=[512, 512],
initializer=tf.contrib.layers.xavier_initializer())
b4 = tf.Variable(tf.random_normal([512]))
L4 = tf.nn.relu(tf.matmul(L3, W4) + b4)
W5 = tf.get_variable("W5_", shape=[512, 10],
initializer=tf.contrib.layers.xavier_initializer())
b5 = tf.Variable(tf.random_normal([10]))
hypothesis = tf.matmul(L4, W5) + b5
In [7]:
# define cost/loss & optimizer
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(
logits=hypothesis, labels=Y))
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)
# initialize
sess = tf.Session()
sess.run(tf.global_variables_initializer())
# train my model
for epoch in range(training_epochs):
avg_cost = 0
total_batch = int(mnist.train.num_examples / batch_size)
for i in range(total_batch):
batch_xs, batch_ys = mnist.train.next_batch(batch_size)
feed_dict = {X: batch_xs, Y: batch_ys}
c, _ = sess.run([cost, optimizer], feed_dict=feed_dict)
avg_cost += c / total_batch
print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.9f}'.format(avg_cost))
print('Learning Finished!')
# Test model and check accuracy
correct_prediction = tf.equal(tf.argmax(hypothesis, 1), tf.argmax(Y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
print('Accuracy:', sess.run(accuracy, feed_dict={
X: mnist.test.images, Y: mnist.test.labels}))
# Get one and predict
r = random.randint(0, mnist.test.num_examples - 1)
print("Label: ", sess.run(tf.argmax(mnist.test.labels[r:r + 1], 1)))
print("Prediction: ", sess.run(
tf.argmax(hypothesis, 1), feed_dict={X: mnist.test.images[r:r + 1]}))
# plt.imshow(mnist.test.images[r:r + 1].
# reshape(28, 28), cmap='Greys', interpolation='nearest')
# plt.show()
더 넓게 더 깊게 쌓았는데 확률이 0.004프로 떨어졌다.
라는 시나리오로 가야 하는데.. 더 높게 나와버렸다..ㅜ
만약 떨어졌다면 왜 이런 일이 발생하는 것일까?
데이터마다 경우가 다르긴 하겠지만, 여기서는 overfitting이다.
그 overffiting을 예방하는 방법중 하나가 drop out이다.
Dropout for MNIST¶
텐서플로우에서 dropout이라는 layer를 하나 더 추가하면된다.
L1 다음에 drop시켜주는 layer 추가. L2 다음에 drop시켜주는 layer 추가. 이런식
얼마나 유지시켜줄지 keep_prop.(test할때는 1로 할 것!)
In [10]:
# dropout (keep_prob) rate 0.7 on training, but should be 1 for testing
keep_prob = tf.placeholder(tf.float32)
# weights & bias for nn layers
# http://stackoverflow.com/questions/33640581/how-to-do-xavier-initialization-on-tensorflow
W1 = tf.get_variable("W1_1", shape=[784, 512],
initializer=tf.contrib.layers.xavier_initializer())
b1 = tf.Variable(tf.random_normal([512]))
L1 = tf.nn.relu(tf.matmul(X, W1) + b1)
L1 = tf.nn.dropout(L1, keep_prob=keep_prob)
W2 = tf.get_variable("W2_1", shape=[512, 512],
initializer=tf.contrib.layers.xavier_initializer())
b2 = tf.Variable(tf.random_normal([512]))
L2 = tf.nn.relu(tf.matmul(L1, W2) + b2)
L2 = tf.nn.dropout(L2, keep_prob=keep_prob)
W3 = tf.get_variable("W3_1", shape=[512, 512],
initializer=tf.contrib.layers.xavier_initializer())
b3 = tf.Variable(tf.random_normal([512]))
L3 = tf.nn.relu(tf.matmul(L2, W3) + b3)
L3 = tf.nn.dropout(L3, keep_prob=keep_prob)
W4 = tf.get_variable("W4_1", shape=[512, 512],
initializer=tf.contrib.layers.xavier_initializer())
b4 = tf.Variable(tf.random_normal([512]))
L4 = tf.nn.relu(tf.matmul(L3, W4) + b4)
L4 = tf.nn.dropout(L4, keep_prob=keep_prob)
W5 = tf.get_variable("W5_1", shape=[512, 10],
initializer=tf.contrib.layers.xavier_initializer())
b5 = tf.Variable(tf.random_normal([10]))
hypothesis = tf.matmul(L4, W5) + b5
In [11]:
# define cost/loss & optimizer
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(
logits=hypothesis, labels=Y))
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)
# initialize
sess = tf.Session()
sess.run(tf.global_variables_initializer())
# train my model
for epoch in range(training_epochs):
avg_cost = 0
for i in range(total_batch):
batch_xs, batch_ys = mnist.train.next_batch(batch_size)
feed_dict = {X: batch_xs, Y: batch_ys, keep_prob: 0.7}
c, _ = sess.run([cost, optimizer], feed_dict=feed_dict)
avg_cost += c / total_batch
print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.9f}'.format(avg_cost))
print('Learning Finished!')
# Test model and check accuracy
correct_prediction = tf.equal(tf.argmax(hypothesis, 1), tf.argmax(Y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
print('Accuracy:', sess.run(accuracy, feed_dict={
X: mnist.test.images, Y: mnist.test.labels, keep_prob: 1}))
# Get one and predict
r = random.randint(0, mnist.test.num_examples - 1)
print("Label: ", sess.run(tf.argmax(mnist.test.labels[r:r + 1], 1)))
print("Prediction: ", sess.run(
tf.argmax(hypothesis, 1), feed_dict={X: mnist.test.images[r:r + 1], keep_prob: 1}))
# plt.imshow(mnist.test.images[r:r + 1].
# reshape(28, 28), cmap='Greys', interpolation='nearest')
# plt.show()
98%가 넘었다. 엄청난 결과다.!
Optimizers¶
train = tf.train.GradientOptimizer(learning_rate=0.1).minimize(cost)
- tf.train.AdadeltaOptimizer
- tf.train.AdagradOptimizer
- tf.train.AdagradDAOptimizer
- tf.train.MomentumOptimizer
- tf.train.AdamOptimizer
- tf.train.FtrlOptimizer
- tf.train.ProximalGradientDescentOptimizer
- tf.train.ProximalAdagradOptimizer
- tf.train.RMSPropOptimizer
Optimizer 종류는 여러가지가 있고 어떤것이 학습이 잘 되는지 테스트 해볼것.
http://www.denizyuret.com/2015/03/alec-radfords-animations-for.html
optimizer를 종류별로 테스트 가능
In [12]:
from PIL import Image
Image.open('adam.png')
Out[12]:
cost가 빨리 줄어들수록 해당 데이터에 적절한 optimizer이다.
보다시피 adam이라는 optimizer가 빨리 줄어드는데, 많이 사용하는 optimizer이니 알아두자
Use Adam Optimizer¶
gradiet descent와 이름부분만 바꾸면 사용할 수 있다.
# define cost/loss & optimizer
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(
logits = hypothesis, labels = Y))
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)Exercise: Batch Normalization¶
입력값을 normalize 잘 하는 방법. 많은 사람들이 사용하고 있다. 이것을 한 번 보고 98% 이상 성능을 올릴 수 있는지 연습해보자.
'beginner > 파이썬 딥러닝 기초' 카테고리의 다른 글
| ConvNet Max Pooling (0) | 2019.05.12 |
|---|---|
| ConvNet의 conv 레이어 만들기 (0) | 2019.05.12 |
| 레고처럼 넷트워크 모듈을 마음껏 쌓아 보자 (0) | 2019.05.08 |
| Dropout과 앙상블 (0) | 2019.05.08 |
| Weight 초기화 잘해보자 (0) | 2019.05.08 |