![[논문읽기] 08-1. WGAN MNIST With Keras](https://img1.daumcdn.net/thumb/R750x0/?scode=mtistory2&fname=https%3A%2F%2Ft1.daumcdn.net%2Fcfile%2Ftistory%2F9990403C5D13207B1A)
728x90
반응형
In [63]:
from keras.models import *
from keras.layers import *
from keras.optimizers import*
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt
import cv2,PIL
In [7]:
(X_train,y_train),(X_test,y_test) = tf.keras.datasets.mnist.load_data()
In [ ]:
X_train = X_train.astype(np.float32).reshape(-1, 28,28) / 255.0
X_test = X_test.astype(np.float32).reshape(-1, 28,28) / 255.0
y_train = y_train.astype(np.int32)
y_test = y_test.astype(np.int32)
In [9]:
X_train = X_train[5000:]
X_valid = X_train[:5000]
y_train = y_train[5000:]
y_valid = y_train[:5000]
In [16]:
# Prints image
plt.rcParams['font.family'] = 'NanumBarunGothic'
plt.imshow(X_train[1])
plt.title("The number is : {0}".format(y_train[1]))
plt.show()
In [1]:
from __future__ import print_function, division
from keras.datasets import mnist
from keras.layers import Input, Dense, Reshape, Flatten, Dropout
from keras.layers import BatchNormalization, Activation, ZeroPadding2D
from keras.layers.advanced_activations import LeakyReLU
from keras.layers.convolutional import UpSampling2D, Conv2D
from keras.models import Sequential, Model
from keras.optimizers import RMSprop
import keras.backend as K
import matplotlib.pyplot as plt
import sys
import numpy as np
class WGAN():
def __init__(self):
self.img_rows = 28
self.img_cols = 28
self.channels = 1
self.img_shape = (self.img_rows, self.img_cols, self.channels)
self.latent_dim = 100
# Following parameter and optimizer set as recommended in paper
self.n_critic = 5
self.clip_value = 0.01
optimizer = RMSprop(lr=0.00005)
# Build and compile the critic
self.critic = self.build_critic()
self.critic.compile(loss=self.wasserstein_loss,
optimizer=optimizer,
metrics=['accuracy'])
# Build the generator
self.generator = self.build_generator()
# The generator takes noise as input and generated imgs
z = Input(shape=(self.latent_dim,))
img = self.generator(z)
# For the combined model we will only train the generator
self.critic.trainable = False
# The critic takes generated images as input and determines validity
valid = self.critic(img)
# The combined model (stacked generator and critic)
self.combined = Model(z, valid)
self.combined.compile(loss=self.wasserstein_loss,
optimizer=optimizer,
metrics=['accuracy'])
def wasserstein_loss(self, y_true, y_pred):
return K.mean(y_true * y_pred)
def build_generator(self):
model = Sequential()
model.add(Dense(128 * 7 * 7, activation="relu", input_dim=self.latent_dim))
model.add(Reshape((7, 7, 128)))
model.add(UpSampling2D())
model.add(Conv2D(128, kernel_size=4, padding="same"))
model.add(BatchNormalization(momentum=0.8))
model.add(Activation("relu"))
model.add(UpSampling2D())
model.add(Conv2D(64, kernel_size=4, padding="same"))
model.add(BatchNormalization(momentum=0.8))
model.add(Activation("relu"))
model.add(Conv2D(self.channels, kernel_size=4, padding="same"))
model.add(Activation("tanh"))
model.summary()
noise = Input(shape=(self.latent_dim,))
img = model(noise)
return Model(noise, img)
def build_critic(self):
model = Sequential()
model.add(Conv2D(16, kernel_size=3, strides=2, input_shape=self.img_shape, padding="same"))
model.add(LeakyReLU(alpha=0.2))
model.add(Dropout(0.25))
model.add(Conv2D(32, kernel_size=3, strides=2, padding="same"))
model.add(ZeroPadding2D(padding=((0,1),(0,1))))
model.add(BatchNormalization(momentum=0.8))
model.add(LeakyReLU(alpha=0.2))
model.add(Dropout(0.25))
model.add(Conv2D(64, kernel_size=3, strides=2, padding="same"))
model.add(BatchNormalization(momentum=0.8))
model.add(LeakyReLU(alpha=0.2))
model.add(Dropout(0.25))
model.add(Conv2D(128, kernel_size=3, strides=1, padding="same"))
model.add(BatchNormalization(momentum=0.8))
model.add(LeakyReLU(alpha=0.2))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(1))
model.summary()
img = Input(shape=self.img_shape)
validity = model(img)
return Model(img, validity)
def train(self, epochs, batch_size=128, sample_interval=50):
# Load the dataset
(X_train, _), (_, _) = mnist.load_data()
# Rescale -1 to 1
X_train = (X_train.astype(np.float32) - 127.5) / 127.5
X_train = np.expand_dims(X_train, axis=3)
# Adversarial ground truths
valid = -np.ones((batch_size, 1))
fake = np.ones((batch_size, 1))
for epoch in range(epochs):
for _ in range(self.n_critic):
# ---------------------
# Train Discriminator
# ---------------------
# Select a random batch of images
idx = np.random.randint(0, X_train.shape[0], batch_size)
imgs = X_train[idx]
# Sample noise as generator input
noise = np.random.normal(0, 1, (batch_size, self.latent_dim))
# Generate a batch of new images
gen_imgs = self.generator.predict(noise)
# Train the critic
d_loss_real = self.critic.train_on_batch(imgs, valid)
d_loss_fake = self.critic.train_on_batch(gen_imgs, fake)
d_loss = 0.5 * np.add(d_loss_fake, d_loss_real)
# Clip critic weights
for l in self.critic.layers:
weights = l.get_weights()
weights = [np.clip(w, -self.clip_value, self.clip_value) for w in weights]
l.set_weights(weights)
# ---------------------
# Train Generator
# ---------------------
g_loss = self.combined.train_on_batch(noise, valid)
# Plot the progress
print ("%d [D loss: %f] [G loss: %f]" % (epoch, 1 - d_loss[0], 1 - g_loss[0]))
# If at save interval => save generated image samples
if epoch % sample_interval == 0:
self.sample_images(epoch)
def sample_images(self, epoch):
r, c = 5, 5
noise = np.random.normal(0, 1, (r * c, self.latent_dim))
gen_imgs = self.generator.predict(noise)
# Rescale images 0 - 1
gen_imgs = 0.5 * gen_imgs + 0.5
fig, axs = plt.subplots(r, c)
cnt = 0
for i in range(r):
for j in range(c):
axs[i,j].imshow(gen_imgs[cnt, :,:,0], cmap='gray')
axs[i,j].axis('off')
cnt += 1
fig.savefig("images/mnist_%d.png" % epoch)
plt.close()
if __name__ == '__main__':
wgan = WGAN()
wgan.train(epochs=4000, batch_size=32, sample_interval=50)
In [177]:
import os
path2 = os.getcwd()
path2
Out[177]:
In [180]:
generated_image_array = [imageio.imread(generated_image) for generated_image in f]
imageio.mimsave(path + 'WGAN_MNIST2.gif', generated_image_array, fps=5)
WGAN을 통해 생성되는 과정을 표시한 것이다.
DCGAN보다 더 선명한 이미지가 나오지 않는 것을 보니 모델을 다시 확인해봐야 할 듯 하다.
728x90
반응형