def result(sess, teX, test_gray, path):
loss, arr = sess.run([cost, X_],
feed_dict={X: teX[0:batch_size],
x_gray: test_gray[0:batch_size],
dropout_conv: 1.0, dropout_fc: 1.0})
NNutils.create_image(arr, path)
def model_fc(self,
z,
layer=[2 * 2 * 256, 2 * 2 * 128, 2 * 2 * 64, 2 * 2 * 16, 10]):
output = z
layer_num = 0
with tf.name_scope('reshape'):
reshape_size = 2 * 2 * 256
output = tf.reshape(output, [-1, reshape_size])
for i in range(3):
layer_num += 1
with tf.variable_scope('fc' + str(layer_num)):
output = NNutils.create_layer('fc',
output,
layer[layer_num],
var_list=self.train_list,
dropout=self.dropout_normal)
layer_num += 1
with tf.variable_scope('fc' + str(layer_num)):
output = NNutils.create_layer('fc',
output,
layer[layer_num],
var_list=self.train_list,
activation='none')
y = output
return y
def run(self, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
# MNIST 데이터
dataset = mnist.read_data_sets("MNIST_data/", one_hot=True)
train_data, train_label, test_data, test_label = dataset.train.images, dataset.train.labels, \
dataset.test.images, dataset.test.labels
train_data = train_data.reshape(
-1, self.time_step,
int(28 * 28 /
self.time_step)) #TF의 RNN은 입력데이터의 rank가 높아야함(열이 많아야 함)
test_data = test_data.reshape(
-1, self.time_step,
int(28 * 28 / self.time_step)) #mnist 데이터는 사실 적절하진 못함
test_indices = np.arange(len(train_data))
np.random.shuffle(test_indices)
test_indices = test_indices[0:10000]
path = "LSTM_mnist/" + str(step_limit) + ""
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, train_data.shape[0], self.batch_size),
range(self.batch_size, train_data.shape[0],
self.batch_size)):
summary, \
_, loss, \
step = sess.run([merged,
self.training, self.cost,
self.global_step],
feed_dict={self.x: train_data[start:end],
self.y: train_label[start:end],
self.dropout_conv: 1.0,
self.dropout_normal: 0.5})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, \
loss, \
accuracy = sess.run([merged, self.cost, self.accuracy],
feed_dict={self.x: test_data,
self.y: test_label,
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
writer_test.add_summary(summary, step)
print("test results : ", accuracy, loss)
saver.save(sess, path + "/model.ckpt", step)
def run(self, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = mnist.read_data_sets("MNIST_data/", one_hot=True)
train_data, train_label, test_data, test_label = dataset.train.images, dataset.train.labels, \
dataset.test.images, dataset.test.labels
# 이 데이터셋은 기본적으로 0~1사이 값으로 정규화되어있어 이미지를 볼 수 없다. => 0~255로 변경
train_data = train_data * 255
test_data = test_data * 255
test_indices = np.arange(len(test_data))
np.random.shuffle(test_indices)
test_indices = test_indices[0:1000]
path = "AE/" + str(step_limit)
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, len(train_data), self.batch_size),
range(self.batch_size, len(train_data),
self.batch_size)):
summary, \
_, loss, \
_, \
step = sess.run([merged,
self.training, self.cost,
self.training_fc,
self.global_step],
feed_dict={self.x: train_data[start:end],
self.y: train_label[start:end],
self.dropout_conv: 0.8,
self.dropout_normal: 0.5})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, loss, loss_fc, im = sess.run(
[merged, self.cost, self.cost_fc, self.im],
feed_dict={
self.x: test_data,
self.y: test_label,
self.dropout_conv: 1.0,
self.dropout_normal: 1.0
})
writer_test.add_summary(summary, step)
print("test results : ", loss, loss_fc)
saver.save(sess, path + "/model.ckpt", step)
def run(self, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = mnist.read_data_sets(one_hot=True)
train_data, train_label, test_data, test_label = dataset.train.images, dataset.train.labels, \
dataset.test.images, dataset.test.labels
test_indices = np.arange(len(test_data))
np.random.shuffle(test_indices)
test_indices = test_indices[0:1000]
name = self.info()
path = "mnist/" + str(step_limit) + name
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, len(train_data), self.batch_size),
range(self.batch_size, len(train_data),
self.batch_size)):
summary, \
_, loss, \
step = sess.run([merged,
self.training, self.cost,
self.global_step],
feed_dict={self.x: train_data[start:end],
self.y: train_label[start:end],
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, \
loss, \
accuracy = sess.run([merged, self.cost, self.accuracy],
feed_dict={self.x: test_data,
self.y: test_label,
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
writer_test.add_summary(summary, step)
print("test results : ", accuracy, loss)
saver.save(
sess,
path + "/" + name + ".ckpt",
step,
)
def run(self, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
#cifar-10 데이터(32*32*3)을 받아오고, fc로 할 수 있도록 형태를 변환한다.(32*32*3 => 3072)
dataset = cifar.Cifar()
train_data, train_label, test_data, test_label = dataset.getdata()
train_data = train_data.reshape(-1, 3072)
test_data = test_data.reshape(-1, 3072)
test_indices = np.arange(len(test_data))
np.random.shuffle(test_indices)
test_indices = test_indices[0:1000]
name = self.info()
path = "cifar/" + str(step_limit) + name
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, len(train_data), self.batch_size),
range(self.batch_size, len(train_data),
self.batch_size)):
summary, \
_, loss, \
step = sess.run([merged,
self.training, self.cost,
self.global_step],
feed_dict={self.x: train_data[start:end],
self.y: train_label[start:end],
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, \
loss, \
accuracy = sess.run([merged, self.cost, self.accuracy],
feed_dict={self.x: test_data,
self.y: test_label,
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
writer_test.add_summary(summary, step)
print("test results : ", accuracy, loss)
saver.save(sess, path + "/" + name + ".ckpt", step)
def run(self, step_limit):
self.train()
rcv1 = fetch_rcv1(subset='train')
train_data = rcv1.data
train_label = rcv1.target
rcv1 = fetch_rcv1(subset='test', random_state=1)
test_data = rcv1.data
test_label = rcv1.target
with tf.Session() as sess:
tf.global_variables_initializer().run()
path = "LSTM/" + str(step_limit) + "rcv1"
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, train_data.shape[0], self.batch_size),
range(self.batch_size, train_data.shape[0],
self.batch_size)):
data = scipy.sparse.coo_matrix(train_data[start:end])
label = scipy.sparse.coo_matrix(train_label[start:end])
indices = np.array([data.row, data.col]).T
summary, \
_, loss, \
step = sess.run([merged,
self.training, self.cost,
self.global_step],
feed_dict={self.x: (indices, data.data, data.shape),
self.y: (indices, label.data, label.shape),
self.dropout_conv: 1.0, self.dropout_normal: 1.0})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, \
loss, \
accuracy = sess.run([merged, self.cost, self.accuracy],
feed_dict={self.x: test_data[0:1000],
self.y: test_label[0:1000],
self.dropout_conv: 1.0, self.dropout_normal: 1.0})
writer_test.add_summary(summary, step)
print("test results : ", accuracy, loss)
def run(self, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = svhn.SVHN()
train_data, train_label = dataset.get_trainset()
test_data, test_label = dataset.get_testset()
train_data = train_data.reshape(-1, 3072)
test_data = test_data.reshape(-1, 3072)
test_indices = np.arange(len(test_data))
np.random.shuffle(test_indices)
test_indices = test_indices[0:1000]
path = "mlp/" + str(step_limit) + "SvhnMlpRelu" + str(
len(self.layers))
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, len(train_data), self.batch_size),
range(self.batch_size, len(train_data),
self.batch_size)):
summary, \
_, loss, \
step = sess.run([merged,
self.training, self.cost,
self.global_step],
feed_dict={self.x: train_data[start:end],
self.y: train_label[start:end],
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, \
loss, \
accuracy = sess.run([merged, self.cost, self.accuracy],
feed_dict={self.x: test_data,
self.y: test_label,
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
writer_test.add_summary(summary, step)
print("test results : ", accuracy, loss)
saver.save(sess, path + "/model.ckpt", step)
def run(epochs):
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = Imagenet.Cifar()
trX, trY, teX, teY = dataset.getdata()
filetime = datetime.now().strftime("%Y_%m_%d_%H_%M_%S")
#path = "Networkfile/" + "convAENN_noise"
path = "Networkfile/convAENN" + filetime
saver = NNutils.save(path, sess)
writer, merged = NNutils.graph(path, sess)
test_indices = np.arange(len(teX))
np.random.shuffle(test_indices)
test_indices = test_indices[0:(batch_size)]
st_time = datetime.now()
for i in range(epochs):
print(i, st_time)
for start, end in zip(range(0, len(trX), batch_size),
range(batch_size, len(trX), batch_size)):
summary, _, loss_nn, loss_ae, learning_rate, step = sess.run(
[merged, trainop, cost_NN, cost_AE, lr, global_step],
feed_dict={
X: trX[start:end],
Y: trY[start:end],
dropout_conv: 0.5,
dropout_fc: 0.5
})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss_nn, loss_ae,
learning_rate)
loss_nn, loss_ae, accuracy = sess.run(
[cost_NN, cost_AE, acc_op],
feed_dict={
X: teX[test_indices],
Y: teY[test_indices],
dropout_conv: 1.0,
dropout_fc: 1.0
})
print("test results : ", accuracy, loss_nn, loss_ae)
saver.save(sess, path + "/model.ckpt", step)
# im = im.astype('uint8')
# im = Image.fromarray(im[0])
# im.save('convAENN.jpg')
end_time = datetime.now()
print("걸린 시간 = ", end_time - st_time)
def run(self, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = cifar.Cifar()
train_data, train_label, test_data, test_label = dataset.getdata()
test_indices = np.arange(len(test_data))
np.random.shuffle(test_indices)
test_indices = test_indices[0:1000]
path = "ConvAE_2dim/50000"
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, len(train_data), self.batch_size),
range(self.batch_size, len(train_data),
self.batch_size)):
summary, \
_, \
loss, \
step = sess.run([merged,
self.training,
self.cost,
self.global_step],
feed_dict={self.x: train_data[start:end],
self.y: train_label[start:end],
self.dropout_conv: 0.8,
self.dropout_normal: 0.5})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, loss = sess.run(
[merged, self.cost],
feed_dict={
self.x: test_data[test_indices],
self.y: test_label[test_indices],
self.dropout_conv: 1.0,
self.dropout_normal: 1.0
})
writer_test.add_summary(summary, step)
print("test results : ", loss)
saver.save(sess, path + "/model.ckpt", step)
def select_dataset(name):
x_size, y_size, train_data, train_label, test_data, test_label = 0, 0, [], [] ,[] ,[] #초기화
if name == 'cifar':
dataset = cifar.CIFAR()
train_data, train_label, test_data, test_label = dataset.getdata()
train_data = train_data.reshape(-1, 3072)
test_data = test_data.reshape(-1, 3072)
x_size = 3072
y_size = 10
elif name == 'svhn':
dataset = svhn.SVHN()
train_data, train_label = dataset.get_trainset()
test_data, test_label = dataset.get_testset()
train_data = train_data.reshape(-1, 3072)
test_data = test_data.reshape(-1, 3072)
x_size = 3072
y_size = 10
elif name == 'mnist':
dataset = mnist.read_data_sets(flags.MNIST_DIR, one_hot=True)
train_data, train_label, test_data, test_label = dataset.train.images, dataset.train.labels, \
dataset.test.images, dataset.test.labels
x_size = 784
y_size = 10
elif name == 'news':
trainset = fetch_20newsgroups(data_home=flags.NEWS_DIR, subset='train')
testset = fetch_20newsgroups(data_home=flags.NEWS_DIR, subset='test')
vectorizer = TfidfVectorizer(analyzer='word', max_features=3072)
vectorizer.fit(trainset.data)
train_data = vectorizer.transform(trainset.data)
train_data = csr_matrix.todense(train_data)
train_label = trainset.target
train_label = NNutils.onehot(train_label, 20, list=True)
# print(train_label.shape)
test_data = vectorizer.transform(testset.data)
test_data = csr_matrix.todense(test_data)
test_label = testset.target
test_label = NNutils.onehot(test_label, 20, list=True)
x_size = 3072
y_size = 20
return Dataset(name, x_size, y_size, train_data, train_label, test_data,
test_label)
def run(self, dataset, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
name = self.info()
path = dataset.name + "/" + str(step_limit) + name
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
step_saved = step
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, len(dataset.train_data), self.batch_size),
range(self.batch_size, len(dataset.train_data),
self.batch_size)):
summary, \
_, loss, \
step = sess.run([merged,
self.opt, self.loss,
self.global_step],
feed_dict={self.x: dataset.train_data[start:end],
self.y: dataset.train_label[start:end],
self.training: True})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, \
loss, \
accuracy = sess.run([merged, self.loss, self.accuracy],
feed_dict={self.x: dataset.test_data,
self.y: dataset.test_label,
self.training: False})
writer_test.add_summary(summary, step)
print("test results : ", accuracy, loss)
if step - step_saved > 1000:
saver.save(sess, path + "/" + name + ".ckpt", step)
step_saved = step
def model_fc(self, z, layer=[7 * 7, 6 * 6, 5 * 5, 10]):
output = z
layer_num = 0
for i in range(2):
layer_num += 1
with tf.variable_scope('fc' + str(layer_num)):
output = NNutils.create_layer('fc',
output,
layer[layer_num],
var_list=self.train_list)
# output = tf.nn.dropout(output, self.dropout_conv)
layer_num += 1
with tf.variable_scope('fc' + str(layer_num)):
output = NNutils.create_layer('fc',
output,
layer[layer_num],
var_list=self.train_list,
activation='none')
y = output
return y
def run(self, epochs):
self.train_unsuper()
#self.train_super()
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = Imagenet.Cifar()
trX, trY, teX, teY = dataset.getdata()
filetime = datetime.now().strftime("%Y_%m_%d_%H_%M")
path = "Networkfile/convAE_sep_class"
#path = "Networkfile/convAENN_sep" + filetime
saver = NNutils.save(path, sess)
writer, merged = NNutils.graph(path, sess)
test_indices = np.arange(len(teX))
np.random.shuffle(test_indices)
test_indices = test_indices[0:(self.batch_size)]
st_time = datetime.now()
for i in range(epochs):
print(i, st_time)
for start, end in zip(
range(0, len(trX), self.batch_size),
range(self.batch_size, len(trX), self.batch_size)):
_ = sess.run(self.trainop_unsuper,
feed_dict={
self.x: trX[start:end],
self.dropout_conv: 0.8,
self.dropout_fc: 0.6
})
# graph_def = graph_pb2.GraphDef()
# output_names = ""
# tf.graph_util.convert_variables_to_constants(sess, graph_def, output_names)
#print(cost_ae)
output = sess.run(self.z,
feed_dict={
self.x: trX[start:end],
self.dropout_conv: 0.8,
self.dropout_fc: 0.6
})
z = tf.constant(output)
print(z.shape)
# cost_nn, _ = sess.run([self.cost_super, self.trainop_super],
# feed_dict={self.x_nn: self.z, self.y: trY[start:end],
# self.dropout_conv: 0.8, self.dropout_fc : 0.6})
#
#_, loss_super, step = sess.run([trainop_super, cost_super, global_step], feed_dict={x:trX[0:0], y: trY[start:end]
# ,dropout_conv: 0.8, dropout_fc : 0.6})
# if step % 50 == 0:
# writer.add_summary(summary, step)
# print(step, loss_super, loss_unsuper)
#print(np.shape(trX))
#summary, accuracy, loss = sess.run([merged, acc_op, cost], feed_dict={ X: teX[test_indices], Y: teY[test_indices]})
#print(step, datetime.now(), loss_unsuper, loss_super, learning_rate)
# loss_su, loss_un, accuracy, step = sess.run([cost_super, cost_unsuper, acc_op, global_step], feed_dict={X: teX[test_indices], Y: teY[test_indices],
# dropout_conv : 1.0, dropout_fc : 1.0})
# print("test results : ", accuracy, loss_super, loss_unsuper)
# saver.save(sess, path + "/model.ckpt", step)
end_time = datetime.now()
print("걸린 시간 = ", end_time - st_time)
def run(epochs):
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = Imagenet.Cifar()
trX, trY, teX, teY = dataset.getdata()
filetime = datetime.now().strftime("%Y_%m_%d_%H_%M_%S")
#path = "Networkfile/convAE_sep" + "2017_03_27_20_32"
path = "Networkfile/convAENN_sep" + filetime
saver = NNutils.save(path, sess)
writer, merged = NNutils.graph(path, sess)
test_indices = np.arange(len(teX))
np.random.shuffle(test_indices)
test_indices = test_indices[0:(batch_size)]
#run_options = tf.RunOptions(trace_level=tf.RunOptions.FULL_TRACE)
#run_metadata = tf.RunMetadata()
st_time = datetime.now()
for i in range(epochs):
print(i, st_time)
for start, end in zip(range(0, len(trX), batch_size),
range(batch_size, len(trX), batch_size)):
summary, _, _, cost_ae, cost_nn, step = sess.run(
[
merged, trainop_unsuper, trainop_super, cost_unsuper,
cost_super, global_step
],
feed_dict={
x: trX[start:end],
y: trY[start:end],
dropout_conv: 1.0,
dropout_fc: 0.8
})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), cost_nn, cost_ae)
accuracy, cost_nn, cost_ae = sess.run(
[acc_op, cost_super, cost_unsuper],
feed_dict={
x: teX[test_indices],
y: teY[test_indices],
dropout_conv: 1.0,
dropout_fc: 1.0
})
saver.save(sess, path + "/model.ckpt", step)
print("test results : ", accuracy, cost_nn, cost_ae)
# _, loss_super, step = sess.run([trainop_super, cost_super, global_step], feed_dict={y: trY[start:end]
# ,dropout_conv: 0.8, dropout_fc : 0.6})
#tf.graph_util.convert_variables_to_constants(sess, )
# if step % 50 == 0:
# writer.add_summary(summary, step)
# print(step, loss_super, loss_unsuper)
#print(np.shape(trX))
#summary, accuracy, loss = sess.run([merged, acc_op, cost], feed_dict={ X: teX[test_indices], Y: teY[test_indices]})
#print(step, datetime.now(), loss_unsuper, loss_super, learning_rate)
# loss_su, loss_un, accuracy, step = sess.run([cost_super, cost_unsuper, acc_op, global_step], feed_dict={X: teX[test_indices], Y: teY[test_indices],
# dropout_conv : 1.0, dropout_fc : 1.0})
# print("test results : ", accuracy, loss_super, loss_unsuper)
# saver.save(sess, path + "/model.ckpt", step)
end_time = datetime.now()
print("걸린 시간 = ", end_time - st_time)
def model(self, x, layer):
image_width = 32
output = x
#first image
layer_num = 0
output1 = output
with tf.variable_scope('1conv' + str(layer_num)):
output1 = NNutils.create_layer('conv', output1, layer[layer_num])
output1 = tf.nn.max_pool(output1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
layer_num += 1
with tf.variable_scope('1conv' + str(layer_num)):
output1 = NNutils.create_layer('conv', output1, layer[layer_num])
output1 = tf.nn.max_pool(output1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
layer_num += 1
with tf.variable_scope('1conv' + str(layer_num)):
output1 = NNutils.create_layer('conv', output1, layer[layer_num])
output1 = tf.nn.max_pool(output1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
layer_num += 1
with tf.variable_scope('1conv' + str(layer_num)):
output1 = NNutils.create_layer('conv', output1, layer[layer_num])
output1 = tf.nn.max_pool(output1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
layer_num += 1
with tf.name_scope('reshape'):
output1 = tf.reshape(output1, [-1, image_width * image_width * layer[3]])
#second image
layer_num = 0
output2 = output
with tf.variable_scope('2conv' + str(layer_num)):
output2 = NNutils.create_layer('conv', output2, layer[layer_num])
output2 = tf.nn.max_pool(output2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
layer_num += 1
with tf.variable_scope('2conv' + str(layer_num)):
output2 = NNutils.create_layer('conv', output2, layer[layer_num])
output2 = tf.nn.max_pool(output2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
layer_num += 1
with tf.variable_scope('2conv' + str(layer_num)):
output2 = NNutils.create_layer('conv', output2, layer[layer_num])
output2 = tf.nn.max_pool(output2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
layer_num += 1
with tf.variable_scope('2conv' + str(layer_num)):
output2 = NNutils.create_layer('conv', output2, layer[layer_num])
output2 = tf.nn.max_pool(output2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
layer_num += 1
with tf.name_scope('reshape'):
output2 = tf.reshape(output2, [-1, image_width * image_width * layer[3]])
#summation layer
with tf.name_scope('concat'):
output = tf.concat([output1, output2], 0)
with tf.variable_scope('normal' + str(layer_num)):
output = NNutils.create_layer('normal', output, layer[layer_num])
layer_num += 1
with tf.variable_scope('normal' + str(layer_num)):
output = NNutils.create_layer('normal', output, layer[layer_num], activation='none')
layer_num += 1
y = output
return y
def run(self, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = input_data.read_data_sets("MNIST_data/", one_hot=True)
train_data, train_label, test_data, test_label = dataset.train.images, dataset.train.labels, \
dataset.test.images, dataset.test.labels
test_indices = np.arange(len(test_data))
np.random.shuffle(test_indices)
test_indices = test_indices[0:1000]
path = "mlp/" + str(step_limit) + "expand"
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, len(train_data), self.batch_size),
range(self.batch_size, len(train_data),
self.batch_size)):
summary, \
_, loss, \
step = sess.run([merged,
self.training, self.cost,
self.global_step],
feed_dict={self.x: train_data[start:end],
self.y: train_label[start:end],
self.dropout_conv: 1.0,
self.dropout_normal: 0.5})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
if step == 1000:
graph = tf.get_default_graph()
old_w = graph.get_tensor_by_name('fc1/fc:0')
exp_w = tf.constant(0,
dtype=tf.float32,
shape=[784, 100],
name='new_w')
old_w = tf.expand_dims(old_w, 1)
graph = tf.get_default_graph()
old_w = graph.get_tensor_by_name('fc1/fc:0')
print(old_w)
# w = tf.assign(old_w, )
summary, \
loss, \
accuracy = sess.run([merged, self.cost, self.accuracy],
feed_dict={self.x: test_data,
self.y: test_label,
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
writer_test.add_summary(summary, step)
print("test results : ", accuracy, loss)
def run(self, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
# dataset = fetch_20newsgroups(data_home="C:/Projects/Data/20newsgroups", subset='all')
trainset = fetch_20newsgroups(data_home=flags.NEWS_DIR,
subset='train')
testset = fetch_20newsgroups(data_home=flags.NEWS_DIR,
subset='test')
vectorizer = TfidfVectorizer(analyzer='word', max_features=3072)
vectorizer.fit(trainset.data)
train_data = vectorizer.transform(trainset.data)
train_data = csr_matrix.todense(train_data)
train_label = trainset.target
train_label = NNutils.onehot(20, train_label) #NNutil 바뀐 이후 수정안됨
# print(train_label.shape)
test_data = vectorizer.transform(testset.data)
test_data = csr_matrix.todense(test_data)
test_label = testset.target
test_label = NNutils.onehot(20, test_label)
# print(test_label.shape)
# dataset = mnist.read_data_sets("C:/Projects/Data/MNIST_data", one_hot=True)
# train_data, train_label, test_data, test_label = dataset.train.images, dataset.train.labels, \
# dataset.test.images, dataset.test.labels
name = self.info()
path = "news/" + str(step_limit) + name
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(
range(0, len(train_data), self.batch_size),
range(self.batch_size, len(train_data),
self.batch_size)):
summary, \
_, loss, \
step = sess.run([merged,
self.training, self.loss,
self.global_step],
feed_dict={self.x: train_data[start:end],
self.y: train_label[start:end],
self.droprate_ph: self.droprate_train})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, \
loss, \
accuracy = sess.run([merged, self.loss, self.accuracy],
feed_dict={self.x: test_data,
self.y: test_label,
self.droprate_ph: self.droprate_test})
writer_test.add_summary(summary, step)
print("test results : ", accuracy, loss)
saver.save(sess, path + "/" + name + ".ckpt", step)
def run(self, step_limit):
self.train()
with tf.Session() as sess:
tf.global_variables_initializer().run()
# dataset = fetch_20newsgroups(data_home="C:/Projects/Data/20newsgroups", subset='all')
trainset = fetch_20newsgroups(data_home="C:/Projects/Data/20newsgroups", subset='train')
testset = fetch_20newsgroups(data_home="C:/Projects/Data/20newsgroups", subset='test')
vectorizer = TfidfVectorizer(analyzer='word', max_features=3072)
vectorizer.fit(trainset.data)
train_data = vectorizer.transform(trainset.data)
train_data = csr_matrix.todense(train_data)
train_label = trainset.target
train_label = NNutils.onehot(20, train_label)
# print(train_label.shape)
test_data = vectorizer.transform(testset.data)
test_data = csr_matrix.todense(test_data)
test_label = testset.target
test_label = NNutils.onehot(20, test_label)
# test_indices = np.arange(len(test_data))
# np.random.shuffle(test_indices)
# test_indices = test_indices[0:1000]
name = self.info()
path = "news/" + str(step_limit) + name
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
step = sess.run(self.global_step)
while step < step_limit:
print("step :", step)
for start, end in zip(range(0, len(train_data), self.batch_size),
range(self.batch_size, len(train_data), self.batch_size)):
summary, \
_, loss, \
step = sess.run([merged,
self.training, self.cost,
self.global_step],
feed_dict={self.x: train_data[start:end],
self.y: train_label[start:end],
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss)
summary, \
loss, \
accuracy = sess.run([merged, self.cost, self.accuracy],
feed_dict={self.x: test_data,
self.y: test_label,
self.dropout_conv: 1.0,
self.dropout_normal: 1.0})
writer_test.add_summary(summary, step)
print("test results : ", accuracy, loss)
saver.save(sess, path + "/" + name + ".ckpt", step)
def run(epochs):
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = Imagenet.Cifar()
trX, trY, teX, teY = dataset.getdata()
print(teY.shape)
filetime = datetime.now().strftime("%Y_%m_%d_%H_%M")
path = "convAE/" + "rgb2rgb"
#path = "Networkfile/convAE" + filetime
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
test_indices = np.arange(len(teX))
np.random.shuffle(test_indices)
test_indices = test_indices[0:batch_size]
st_time = datetime.now()
for i in range(epochs):
print(i, st_time)
for start, end in zip(range(0, len(trX), batch_size), range(batch_size, len(trX), batch_size)):
summary, _,\
loss, learning_rate,\
step = sess.run([merged, trainop,
cost, lr,
global_step],
feed_dict={ X: trX[start:end],
dropout_conv : 0.8, dropout_fc : 0.5})
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss, learning_rate)
loss, results = sess.run([cost, Z], feed_dict={X: teX,
dropout_conv: 1.0,
dropout_fc: 1.0,
})
print("test results : ", loss)
saver.save(sess, path + "/model.ckpt", step)
#
# image = image.astype('uint8')
# im = Image.fromarray(image[0])
# im.show()
# image = teX.astype('uint8')
# im = Image.fromarray(image[0])
# im.show()
end_time = datetime.now()
print("걸린 시간 = ", end_time - st_time)
#평가
# SVM 학습
loss, x_train = sess.run([cost, Z], feed_dict={X: trX,
dropout_conv: 1.0,
dropout_fc: 1.0,
})
x_train = x_train.reshape(len(x_train), -1)
y_train = np.argmax(trY, 1)
print(x_train.shape)
clf = svm.LinearSVC(max_iter=500, random_state=2)
clf.fit(x_train, y_train)
# SVM 예측 정확도 계산
loss, x_test = sess.run([cost, Z], feed_dict={X: teX,
dropout_conv: 1.0,
dropout_fc: 1.0,
})
x_test = x_test.reshape(len(x_test), -1)
y_test = np.argmax(teY, 1)
accuracy = 0
iteration = 50
for i in range(iteration):
print(i)
clf = svm.LinearSVC(max_iter=200)
clf.fit(x_train, y_train)
acc = clf.score(x_test, y_test)
print(acc)
accuracy += acc
accuracy /= iteration
print(accuracy)
def run(epochs):
with tf.Session() as sess:
tf.global_variables_initializer().run()
dataset = Imagenet.Cifar()
trX, trY, teX, teY = dataset.getdata()
train_gray, test_gray = dataset.getgray()
#train_gray = train_gray.reshape(-1, 32, 32, 1)
#test_gray = test_gray.reshape(-1, 32, 32, 1)
filetime = datetime.now().strftime("%Y_%m_%d_%H_%M")
path = "ConvAE/" + "rgb2gray"
#path = "Networkfile/convAE" + filetime
saver = NNutils.save(path, sess)
writer, writer_test, merged = NNutils.graph(path, sess)
test_indices = np.arange(len(teX))
np.random.shuffle(test_indices)
test_indices = test_indices[0:(batch_size)] #deconv때문에 batchsize보다 크면 문제가 발생한다.
st_time = datetime.now()
for i in range(epochs):
print(i, st_time)
for start, end in zip(range(0, len(trX), batch_size), range(batch_size, len(trX), batch_size)):
summary,\
_, loss, \
learning_rate, step = sess.run([merged,
trainop, cost,
# trainop_nn, cost_nn,
lr, global_step],
feed_dict={ X: trX[start:end],
# Y: trY[start:end],
x_gray: train_gray[start:end],
dropout_conv : 0.8, dropout_fc : 1.0})
#print(loss)
if step % 50 == 0:
writer.add_summary(summary, step)
print(step, datetime.now(), loss, learning_rate)
loss, arr = sess.run([cost, X_],
feed_dict={X: teX,
# Y: teY[0:batch_size],
x_gray: test_gray,
dropout_conv : 1.0, dropout_fc : 1.0})
# arr_uint = arr.astype('uint8')
# #arr_uint = arr_uint.reshape(-1, 32 * 32 * 1)
# im = Image.fromarray(arr_uint[0])
# im.show()
print("test results : ", loss)
saver.save(sess, path + "/model.ckpt", step)
end_time = datetime.now()
print("걸린 시간 = ", end_time - st_time)
# 평가
# SVM 학습
x_train = sess.run(Z, feed_dict={X: trX,
x_gray: test_gray,
dropout_conv: 1.0,
dropout_fc: 1.0,
})
x_train = x_train.reshape(len(x_train), -1)
y_train = np.argmax(trY, 1)
print(x_train.shape)
# SVM 예측 정확도 계산
x_test = sess.run(Z, feed_dict={X: teX,
x_gray: test_gray,
dropout_conv: 1.0,
dropout_fc: 1.0,
})
x_test = x_test.reshape(len(x_test), -1)
y_test = np.argmax(teY, 1)
accuracy = 0
iteration = 50
for i in range(iteration):
print(i)
clf = svm.LinearSVC(max_iter=200)
clf.fit(x_train, y_train)
acc = clf.score(x_test, y_test)
print(acc)
accuracy += acc
accuracy /= iteration
print(accuracy)
buffer[i][2] = 1 #점수 줌
if episode % 100 == 0:
env.render()
reward_all += reward
#
for key, action, value in buffer:
mem.add(key, action, value)
# print(mem.buffer)
x = np.array(list(mem.buffer.keys()))
y = np.array(list(mem.buffer.values()))
y = y[:, :, 2]
x = NNutils.onehot(x, env.observation_space.n, list=True)
# print("input & label",x.shape, y.shape)
# print("x, y :", x, y)
_, loss = net.update(sess, x, y)
# print("loss", loss)
graph.append(reward_all)
x = np.array(list(mem.buffer.keys()))
y = np.array(list(mem.buffer.values()))
y = y[:, :, 2]
print(x, y)
print("Success rate: " + str(sum(graph) / num_episodes))
plt.bar(range(len(graph)), graph, color="blue")
plt.show()
import scipy.sparse from sklearn.datasets import fetch_20newsgroups # from sklearn.datasets import fetch_20newsgroups_vectorized from sklearn.feature_extraction.text import TfidfVectorizer # from datetime import datetime #20newsgroups 셋을 불러옴 dataset = fetch_20newsgroups(data_home="C:/YWK/Projects/Python/Data/20newsgroups", subset='all') #이걸로 하지말고 train에 맞춰서 해보자. trainset = fetch_20newsgroups(data_home="C:/YWK/Projects/Python/Data/20newsgroups", subset='train') testset = fetch_20newsgroups(data_home="C:/YWK/Projects/Python/Data/20newsgroups", subset='test') #벡터화 vectorizer = TfidfVectorizer(analyzer='word', max_features=1000) vectorizer.fit(dataset.data) train_data = vectorizer.transform(trainset.data) train_data = scipy.sparse.csr_matrix.todense(train_data) train_label = trainset.target train_label = NNutils.onehot(20, train_label) print(train_data.shape) print(train_data[0]) # test_data = vectorizer.fit_transform(testset.data) # test_data = scipy.sparse.csr_matrix.todense(test_data) # test_label = testset.target # print(test_data[0])
# trX, trY, teX, teY = mnist.train.images, mnist.train.labels, mnist.test.images, mnist.test.labels
dataset = Imagenet.Cifar()
trX, trY, teX, teY = dataset.getdata()
trX = trX.reshape(-1, 32 * 32 * 3)
teX = teX.reshape(-1, 32 * 32 * 3)
print(tf.shape(cost))
with tf.Session() as sess:
tf.global_variables_initializer().run()
batch_size = 128
st_time = datetime.now()
print(st_time)
savepath = "Networkfile/NN"
saver = NNutils.save(savepath, sess)
writer, merge = NNutils.graph("Networkfile/NN", sess)
for i in range(1):
for start, end in zip(range(0, len(trX), batch_size),
range(batch_size, len(trY), batch_size)):
summary, _, loss, step = sess.run(
[merge, train_op, cost, global_step],
feed_dict={
X: trX[start:end],
Y: trY[start:end]
})
print(step, loss)
writer.add_summary(summary, step)
#saver.save(sess, savepath + "/model.ckpt", step)