-
Notifications
You must be signed in to change notification settings - Fork 50
/
simple_conv_gan_EC2.py
137 lines (122 loc) · 4.43 KB
/
simple_conv_gan_EC2.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
# Simple GAN implementation with keras
# adaptation of https://gist.github.com/Newmu/4ee0a712454480df5ee3
import sys
from keras.models import Sequential
from keras.layers.core import Dense,Dropout,Flatten
from keras.layers.convolutional import Convolution1D, UpSampling1D
from keras.layers.normalization import BatchNormalization
from keras.optimizers import SGD,RMSprop
from keras.initializations import normal
import numpy as np
from scipy.stats import gaussian_kde
from scipy.io import wavfile
import theano.tensor as T
import theano
import pydub
batch_size = 32
print "loading data"
f = pydub.AudioSegment.from_mp3('07_-_Brad_Sucks_-_Total_Breakdown.mp3')
data = np.fromstring(f._data, np.int16)
data = data.astype(np.float64).reshape((-1,2))
print data.shape
data = data[:,0]+data[:,1]
#data = data[:,:subsample*int(len(data)/subsample)-1,:]
data -= data.min()
data /= data.max() / 2.
data -= 1.
print data.shape
print "Setting up decoder"
decoder = Sequential()
decoder.add(Convolution1D(4, 256, subsample_length = 4, input_shape=(32768,1), activation='relu'))
decoder.add(Dropout(0.5))
decoder.add(Convolution1D(16, 256, input_dim=4, subsample_length = 16, activation='relu'))
decoder.add(Dropout(0.5))
decoder.add(Flatten())
decoder.add(Dense(1, activation='sigmoid'))
#sgd = SGD(lr=0.01, momentum=0.1)
sgd1 = RMSprop()
decoder.compile(loss='binary_crossentropy', optimizer=sgd1)
print "Setting up generator"
generator = Sequential()
generator.add(Convolution1D(16, 64, activation='relu', input_shape=(2048,16),border_mode='same'))
generator.add(BatchNormalization())
generator.add(UpSampling1D(4))
generator.add(Convolution1D(4, 256, activation='relu',border_mode='same'))
generator.add(BatchNormalization())
generator.add(UpSampling1D(4))
#(((2048+64-1)*4 + 256 - 1)*4 - 2029 + 1)
generator.add(Convolution1D(1, 512, activation='linear',border_mode='same'))
sgd = RMSprop()
generator.compile(loss='binary_crossentropy', optimizer=sgd)
print "Setting up combined net"
gen_dec = Sequential()
gen_dec.add(generator)
decoder.trainable=False
gen_dec.add(decoder)
#def inverse_binary_crossentropy(y_true, y_pred):
# if theano.config.floatX == 'float64':
# epsilon = 1.0e-9
# else:
# epsilon = 1.0e-7
# y_pred = T.clip(y_pred, epsilon, 1.0 - epsilon)
# bce = T.nnet.binary_crossentropy(y_pred, y_true).mean(axis=-1)
# return -bce
#
#gen_dec.compile(loss=inverse_binary_crossentropy, optimizer=sgd)
sgd3 = RMSprop()
gen_dec.compile(loss='binary_crossentropy', optimizer=sgd3)
y_decode = np.ones(2*batch_size)
y_decode[:batch_size] = 0.
y_gen_dec = np.ones(batch_size)
def gaussian_likelihood(X, u=0., s=1.):
return (1./(s*np.sqrt(2*np.pi)))*np.exp(-(((X - u)**2)/(2*s**2)))
#def vis(i):
# s = 1.
# u = 0.
# zs = np.linspace(-1, 1, 500).astype('float32')[:,np.newaxis]
# xs = np.linspace(-5, 5, 500).astype('float32')[:,np.newaxis]
# ps = gaussian_likelihood(xs, 1.)
#
# gs = generator.predict(zs)
# print gs.mean(),gs.std()
# preal = decoder.predict(xs)
# kde = gaussian_kde(gs.flatten())
#
# plt.clf()
# plt.plot(xs, ps, '--', lw=2)
# plt.plot(xs, kde(xs.T), lw=2)
# plt.plot(xs, preal, lw=2)
# plt.xlim([-5., 5.])
# plt.ylim([0., 1.])
# plt.ylabel('Prob')
# plt.xlabel('x')
# plt.legend(['P(data)', 'G(z)', 'D(x)'])
# plt.title('GAN learning gaussian')
# fig.canvas.draw()
# plt.show(block=False)
# if i%100 == 0:
# plt.savefig('current.png')
# plt.pause(0.01)
#fig = plt.figure()
for i in range(1000000):
zmb = np.random.normal(0., 1, size=(batch_size, 2048, 16)).astype('float32')
#xmb = np.random.normal(1., 1, size=(batch_size, 1)).astype('float32')
xmb = np.array([data[n:n+32768] for n in np.random.randint(0,data.shape[0]-32768,batch_size)]).astype('float32')
xmb = xmb[:,:,np.newaxis]
if i % 10 == 0:
err_E = 2
while err_E > 0.9:
r = gen_dec.fit(zmb,y_gen_dec,nb_epoch=1,verbose=0)
err_E = r.totals['loss']/batch_size
print 'E:',err_E
else:
r = decoder.fit(np.vstack([generator.predict(zmb),xmb]),y_decode,nb_epoch=1,verbose=0)
print 'D:',(r.totals['loss']/r.seen)
if i % 100 == 0:
print "saving fakes"
fakes = generator.predict(zmb[:16,:])
wavfile.write('fake_epoch_'+str(i)+'.wav',44100,fakes[0,:])
for n in range(16):
wavfile.write('fake_'+str(n+1)+'.wav',44100,fakes[n,:])
wavfile.write('real_'+str(n+1)+'.wav',44100,xmb[n,:])
# vis(i)