sys.path.insert(0, "./")
import csgan as cs
if os.path.exists('../dataset/'):
prefix = '../dataset/'
elif os.path.exists('../../dataset/'):
prefix = '../../dataset/'
else:
raise Exception('Dataset not found!')
dataset_files = [
'map1n_allz_rtaapixlw_2048_1.fits', 'map1n_allz_rtaapixlw_2048_2.fits',
'map1n_allz_rtaapixlw_2048_3.fits'
]
dp = cs.Data_Provider([prefix + file_name for file_name in dataset_files],
preprocess_mode=2)
batch_size = 16
image_size = 128
gf_dim = 64
df_dim = 64
z_dim = 128
def dpp(n):
return dp(n, image_size).reshape(n, image_size, image_size, 1)
##################################################################
from csgan.ops import lrelu, conv2d, linear
import tensorflow as tf
d =np.sqrt(scharrx**2+scharry**2)
return d
def filt_all(maps,func):
out1 = []
for m in maps:
out1.append(func(m))
return np.stack([maps,np.array(out1)],axis=3)
def func(dt):
return canny(dt,0,'none','sch')
dp = cs.Data_Provider('../../../strings/map1n_allz_rtaapixlw_4096_1b')
ims = 512
import tensorflow as tf
def conv2d(x, W):
return tf.nn.conv2d(input=x, filter=W, strides=[1, 1, 1, 1], padding='SAME')
def avg_pool_2x2(x):
return tf.nn.avg_pool(x, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
def discriminator(x_image, reuse=False, n_channel=2):
with tf.variable_scope('discriminator') as scope:
if (reuse):
tf.get_variable_scope().reuse_variables()
#First Conv and Pool Layers
out1 = []
for m in maps:
out1.append(func(m))
return np.stack([maps, np.array(out1)], axis=3)
def func(dt):
return canny(dt, 0, 'none', 'sch')
file_list = [
'../../dataset/map1n_allz_rtaapixlw_2048_' + str(i) + '.fits'
for i in range(1, 4)
]
dp = cs.Data_Provider(file_list, preprocess_mode=2)
#dt = filt_all(dp(10,128),func)
#dt.shape
#fig,(ax1,ax2)=plt.subplots(1,2,figsize=(8,18))
#ax1.imshow(dt[0,:,:,0])
#ax2.imshow(dt[0,:,:,1])
batch_size = 64
image_size = 256
checkpoint_dir = './checkpoint/' + sys.argv[0][:-3]
sample_dir = './samples/' + sys.argv[0][:-3]
def dpp(n):
# return dp(n,image_size).reshape(n,image_size,image_size,1)
out1 = []
for m in maps:
out1.append(func(m))
return np.stack([maps, np.array(out1)], axis=3)
def func(dt):
return canny(dt, 0, 'none', 'sch')
file_list = [
'../../dataset/map1n_allz_rtaapixlw_2048_' + str(i) + '.fits'
for i in range(1, 2)
]
dp = cs.Data_Provider(file_list)
#dt = filt_all(dp(10,128),func)
#dt.shape
#fig,(ax1,ax2)=plt.subplots(1,2,figsize=(8,18))
#ax1.imshow(dt[0,:,:,0])
#ax2.imshow(dt[0,:,:,1])
batch_size = 512
image_size = 128
gf_dim = 8
z_dim = 4096
#(image_size/16)**2*gf_dim*8==z_dim