import sugartensor as tf from data import SpeechCorpus, voca_size from model import * __author__ = '*****@*****.**' # set log level to debug tf.sg_verbosity(10) # # hyper parameters # batch_size = 16 # total batch size # # inputs # # corpus input tensor data = SpeechCorpus(batch_size=batch_size * tf.sg_gpus()) # mfcc feature of audio inputs = tf.split(data.mfcc, tf.sg_gpus(), axis=0) # target sentence label labels = tf.split(data.label, tf.sg_gpus(), axis=0) # sequence length except zero-padding
import sugartensor as tf
from data import SpeechCorpus, voca_size, index2str
from model import *
import numpy as np
from tqdm import tqdm
from attacks import FastGradientMethod
import pickle
##################################################
## Edited by Jade Huang
##################################################
__author__ = '*****@*****.**'
# set log level to debug
tf.sg_verbosity(10)
# command line argument for set_name
tf.sg_arg_def(set=('valid',
"'train', 'valid', or 'test'. The default is 'valid'"))
tf.sg_arg_def(frac=(
1.0,
"test fraction ratio to whole data set. The default is 1.0(=whole set)"))
#
# hyper parameters
#
# batch size
batch_size = 16
import tensorflow as tf
import sugartensor as sg
from input_data import Surv
from model import ResNet
sg.sg_verbosity(10)
batch_size = 32
data = Surv(batch_size=32)
input_ = data.data
label = data.label
model = ResNet()
inference_fn = model.inference
# logit = model.inference(x)
def my_loss(input_, labels, inference_fn, num_gpu=1):
assert num_gpu >= 0
tower_loss = []
input_batch = tf.split(input_, num_gpu, axis=0)
label_batch = tf.split(labels, num_gpu, axis=0)
for i in range(num_gpu):
with tf.device('/gpu:%d' % i):
with tf.name_scope('gpu_%d' % i):
reuse = False if i == 0 else True
logit = inference_fn(input_batch[i], reuse=reuse)
import sugartensor as tf from model import * from tw_data import TwitData __author__ = '*****@*****.**' # set log level to debug tf.sg_verbosity(1) # # hyper parameters # batch_size = 8 # batch size # # inputs # data = TwitData(batch_size=batch_size) # source, target sentence x, y = data.source, data.target # shift target for training source y_in = tf.concat([tf.zeros((batch_size, 1), tf.sg_intx), y[:, :-1]], axis=1) # vocabulary size voca_size = data.voca_size # make embedding matrix for source and target emb_x = tf.sg_emb(name='emb_x', voca_size=voca_size, dim=latent_dim) emb_y = tf.sg_emb(name='emb_y', voca_size=voca_size, dim=latent_dim)
def main(argv):
# set log level to debug
tf.sg_verbosity(10)
#
# hyper parameters
#
size = 160, 147
batch_size = 1 # batch size
#
# inputs
#
pngName = argv
png = tf.read_file(pngName)
#png.thumbnail(size, Image.ANTIALIAS)
#png = tf.resize(png1, (14,14))
myPNG = tf.image.decode_png(png)
y = convert_image(pngName)
x = tf.reshape(y, [1, 28, 28, 1])
print(x)
# corrupted image
x_small = tf.image.resize_bicubic(x, (14, 14))
x_bicubic = tf.image.resize_bicubic(x_small, (28, 28)).sg_squeeze()
x_nearest = tf.image.resize_images(
x_small, (28, 28),
tf.image.ResizeMethod.NEAREST_NEIGHBOR).sg_squeeze()
#
# create generator
#
# I've used ESPCN scheme
# http://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Shi_Real-Time_Single_Image_CVPR_2016_paper.pdf
#
# generator network
with tf.sg_context(name='generator', act='relu', bn=True):
gen = (x.sg_conv(dim=32).sg_conv().sg_conv(
dim=4, act='sigmoid',
bn=False).sg_periodic_shuffle(factor=2).sg_squeeze())
#
# run generator
#
fileName = "inPython.png"
fig_name = "genImages/" + fileName
#fig_name2 = 'asset/train/sample2.png'
print("start")
with tf.Session() as sess:
with tf.sg_queue_context(sess):
tf.sg_init(sess)
# restore parameters
saver = tf.train.Saver()
#saver.restore(sess, tf.train.latest_checkpoint('asset/train/ckpt'))
saver.restore(
sess, tf.train.latest_checkpoint('python/asset/train/ckpt'))
# run generator
gt, low, bicubic, sr = sess.run(
[x.sg_squeeze(), x_nearest, x_bicubic, gen])
# plot result
#sr[0].thumbnail(size, Image.ANTIALIAS)
plt.figure(figsize=(1, 1))
#plt.set_axis_off()
hr = plt.imshow(sr[0], 'gray')
plt.axis('tight')
plt.axis('off')
#ax.set_axis_off()
#ax.thumbnail(size, Image.ANTIALIAS)
#plt.savefig(fig_name,bbox_inches='tight',pad_inches=0,dpi=600)
plt.savefig(fig_name, dpi=600)
#tf.sg_info('Sample image saved to "%s"' % fig_name)
plt.close()
##print (type (sr[0]))
##sourceImage = Image.fromarray(np.uint8(sr[0])
print("done")
