def __call__(self, shape, name=None):
flat_shape = (shape[0], np.prod(shape[1:]))
a = np_rng.normal(0.0, 1.0, flat_shape)
u, _, v = np.linalg.svd(a, full_matrices=False)
q = u if u.shape == flat_shape else v # pick the one with the correct shape
q = q.reshape(shape)
return sharedX(self.scale * q[:shape[0], :shape[1]], name=name)
def __call__(self, shape, name=None):
flat_shape = (shape[0], np.prod(shape[1:]))
a = np_rng.normal(0.0, 1.0, flat_shape)
u, _, v = np.linalg.svd(a, full_matrices=False)
q = u if u.shape == flat_shape else v # pick the one with the correct shape
q = q.reshape(shape)
return sharedX(self.scale * q[:shape[0], :shape[1]], name=name)
def __call__(self, shape):
if len(shape) == 2:
scale = np.sqrt(2. / shape[0])
elif len(shape) == 4:
scale = np.sqrt(2. / np.prod(shape[1:]))
else:
raise NotImplementedError
return sharedX(np_rng.normal(size=shape, scale=scale))
def __call__(self, shape):
if len(shape) == 2:
scale = np.sqrt(2./shape[0])
elif len(shape) == 4:
scale = np.sqrt(2./np.prod(shape[1:]))
else:
raise NotImplementedError
return sharedX(np_rng.normal(size=shape, scale=scale))
def init(self):
naxes = len(self.out_shape)
if naxes == 2 or naxes == 4:
dim = self.out_shape[1]
elif naxes == 3:
dim = self.out_shape[-1]
else:
raise NotImplementedError
self.g = inits.Constant(c=1.)(dim)
self.b = inits.Constant(c=0.)(dim)
self.u = inits.Constant(c=0.)(dim)
self.s = inits.Constant(c=0.)(dim)
self.n = sharedX(0.)
self.params = [self.g, self.b]
self.other_params = [self.u, self.s, self.n]
def init(self):
naxes = len(self.out_shape)
if naxes == 2 or naxes == 4:
dim = self.out_shape[1]
elif naxes == 3:
dim = self.out_shape[-1]
else:
raise NotImplementedError
self.g = inits.Constant(c=1.)(dim)
self.b = inits.Constant(c=0.)(dim)
self.u = inits.Constant(c=0.)(dim)
self.s = inits.Constant(c=0.)(dim)
self.n = sharedX(0.)
self.params = [self.g, self.b]
self.other_params = [self.u, self.s, self.n]
def __call__(self, vocab, name=None):
t = time()
w2v_vocab = joblib.load(os.path.join(self.data_dir, '3m_w2v_gn_vocab.jl'))
w2v_embed = joblib.load(os.path.join(self.data_dir, '3m_w2v_gn.jl'))
mapping = {}
for i, w in enumerate(w2v_vocab):
w = w.lower()
if w in mapping:
mapping[w].append(i)
else:
mapping[w] = [i]
widxs = []
w2vidxs = []
for i, w in enumerate(vocab):
w = w.replace('`', "'")
if w in mapping:
w2vi = min(mapping[w])
w2vidxs.append(w2vi)
widxs.append(i)
w = np.zeros((len(vocab), w2v_embed.shape[1]))
w[widxs, :] = w2v_embed[w2vidxs, :]/2.
return sharedX(w, name=name)
def __call__(self, vocab, name=None):
t = time()
w2v_vocab = joblib.load(
os.path.join(self.data_dir, '3m_w2v_gn_vocab.jl'))
w2v_embed = joblib.load(os.path.join(self.data_dir, '3m_w2v_gn.jl'))
mapping = {}
for i, w in enumerate(w2v_vocab):
w = w.lower()
if w in mapping:
mapping[w].append(i)
else:
mapping[w] = [i]
widxs = []
w2vidxs = []
for i, w in enumerate(vocab):
w = w.replace('`', "'")
if w in mapping:
w2vi = min(mapping[w])
w2vidxs.append(w2vi)
widxs.append(i)
w = np.zeros((len(vocab), w2v_embed.shape[1]))
w[widxs, :] = w2v_embed[w2vidxs, :] / 2.
return sharedX(w, name=name)
def __call__(self, shape):
return sharedX(np.identity(shape[0]) * self.scale)
def __call__(self, shape):
return sharedX(np.ones(shape) * self.c)
def __call__(self, shape, name=None):
r = np_rng.normal(loc=0, scale=0.01, size=shape)
r = r/np.sqrt(np.sum(r**2))*np.sqrt(shape[1])
return sharedX(r, name=name)
def __call__(self, shape, name=None):
return sharedX(np_rng.normal(loc=self.loc, scale=self.scale, size=shape), name=name)
def __call__(self, shape):
return sharedX(np.ones(shape) * self.c)
def __call__(self, shape, name=None):
r = np_rng.normal(loc=0, scale=0.01, size=shape)
r = r / np.sqrt(np.sum(r**2)) * np.sqrt(shape[1])
return sharedX(r, name=name)
def __call__(self, shape, name=None):
return sharedX(np_rng.normal(loc=self.loc,
scale=self.scale,
size=shape),
name=name)
def __call__(self, shape):
return sharedX(
np_rng.uniform(low=-self.scale, high=self.scale, size=shape))
#get the cost functions from both the real, generated, and discriminator functions d_cost_real = bce(p_real, T.ones(p_real.shape)).mean() d_cost_gen = bce(p_gen, T.zeros(p_gen.shape)).mean() g_cost_d = bce(p_gen, T.ones(p_gen.shape)).mean() #get the cost functions from the discriminator function d_cost = d_cost_real + d_cost_gen g_cost = g_cost_d #create the grand cost function cost = [g_cost, d_cost, d_cost_real, d_cost_gen] #update the weights using the 'adam' method https://arxiv.org/pdf/1412.6980 lr = 0.001 lrt = sharedX(lr) d_updater = updates.Adam(lr=lrt) g_updater = updates.Adam(lr=lrt) d_updates = d_updater(d_params, d_cost) g_updates = g_updater(g_params, g_cost) updates = d_updates + g_updates #get the final variables for both networks, cost, and score (to keep track of things) _train_g = theano.function([X, Z], cost, updates=g_updates) _train_d = theano.function([X, Z], cost, updates=d_updates) _train_both = theano.function([X, Z], cost, updates=updates) _gen = theano.function([Z], gen) _score = theano.function([X], p_real) _cost = theano.function([X, Z], cost)
def __call__(self, shape):
return sharedX(np.identity(shape[0]) * self.scale)
#get the cost functions from both the real, generated, and discriminator functions d_cost_real = bce(p_real, T.ones(p_real.shape)).mean() d_cost_gen = bce(p_gen, T.zeros(p_gen.shape)).mean() g_cost_d = bce(p_gen, T.ones(p_gen.shape)).mean() #get the cost functions from the discriminator function d_cost = d_cost_real + d_cost_gen g_cost = g_cost_d #create the grand cost function cost = [g_cost, d_cost, d_cost_real, d_cost_gen] #update the weights using the 'adam' method https://arxiv.org/pdf/1412.6980 lr = 0.001 lrt = sharedX(lr) d_updater = updates.Adam(lr=lrt) g_updater = updates.Adam(lr=lrt) d_updates = d_updater(d_params, d_cost) g_updates = g_updater(g_params, g_cost) updates = d_updates + g_updates #get the final variables for both networks, cost, and score (to keep track of things) _train_g = theano.function([X, Z], cost, updates=g_updates) _train_d = theano.function([X, Z], cost, updates=d_updates) _train_both = theano.function([X, Z], cost, updates=updates) _gen = theano.function([Z], gen) _score = theano.function([X], p_real) _cost = theano.function([X, Z], cost)
def __call__(self, shape):
return sharedX(np_rng.uniform(low=-self.scale, high=self.scale, size=shape))