def get_updates(self, loss, params):
grads = self.get_gradients(loss, params)
self.updates = [K.update_add(self.iterations, 1)]
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * K.cast(self.iterations,
K.dtype(self.decay))))
t = K.cast(self.iterations, K.floatx()) + 1
lr_t = lr * (K.sqrt(1. - K.pow(self.beta_2, t)) /
(1. - K.pow(self.beta_1, t)))
ms = [K.zeros(K.int_shape(p), dtype=K.dtype(p)) for p in params]
vs = [K.zeros(K.int_shape(p), dtype=K.dtype(p)) for p in params]
self.weights = [self.iterations] + ms + vs
for p, g, m, v in zip(params, grads, ms, vs):
m_t = (self.beta_1 * m) + (1. - self.beta_1) * g
v_t = (self.beta_2 * v) + (1. - self.beta_2) * K.square(g)
p_t = p - lr_t * m_t / (K.sqrt(v_t) + self.epsilon)
self.updates.append(K.update(m, m_t))
self.updates.append(K.update(v, v_t))
new_p = p_t
# Apply constraints.
if getattr(p, 'constraint', None) is not None:
new_p = p.constraint(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
self.updates = [K.update_add(self.iterations, 1)]
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * self.iterations))
t = self.iterations + 1
lr_t = lr / (1. - K.pow(self.beta_1, t))
shapes = [K.int_shape(p) for p in params]
# zero init of 1st moment
ms = [K.zeros(shape) for shape in shapes]
# zero init of exponentially weighted infinity norm
us = [K.zeros(shape) for shape in shapes]
self.weights = [self.iterations] + ms + us
for p, g, m, u in zip(params, grads, ms, us):
m_t = (self.beta_1 * m) + (1. - self.beta_1) * g
u_t = K.maximum(self.beta_2 * u, K.abs(g))
p_t = p - lr_t * m_t / (u_t + self.epsilon)
self.updates.append(K.update(m, m_t))
self.updates.append(K.update(u, u_t))
new_p = p_t
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
shapes = [K.int_shape(p) for p in params]
accumulators = [K.zeros(shape) for shape in shapes]
delta_accumulators = [K.zeros(shape) for shape in shapes]
self.weights = accumulators + delta_accumulators
self.updates = []
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * self.iterations))
self.updates.append(K.update_add(self.iterations, 1))
for p, g, a, d_a in zip(params, grads, accumulators, delta_accumulators):
# update accumulator
new_a = self.rho * a + (1. - self.rho) * K.square(g)
self.updates.append(K.update(a, new_a))
# use the new accumulator and the *old* delta_accumulator
update = g * K.sqrt(d_a + self.epsilon) / K.sqrt(new_a + self.epsilon)
new_p = p - lr * update
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
# update delta_accumulator
new_d_a = self.rho * d_a + (1 - self.rho) * K.square(update)
self.updates.append(K.update(d_a, new_d_a))
return self.updates
def get_updates(self, loss, params):
grads = self.get_gradients(loss, params)
shapes = [K.int_shape(p) for p in params]
accumulators = [K.zeros(shape) for shape in shapes]
delta_accumulators = [K.zeros(shape) for shape in shapes]
self.weights = accumulators + delta_accumulators
self.updates = [K.update_add(self.iterations, 1)]
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * K.cast(self.iterations,
K.dtype(self.decay))))
for p, g, a, d_a in zip(params, grads, accumulators, delta_accumulators):
# update accumulator
new_a = self.rho * a + (1. - self.rho) * K.square(g)
self.updates.append(K.update(a, new_a))
# use the new accumulator and the *old* delta_accumulator
update = g * K.sqrt(d_a + self.epsilon) / K.sqrt(new_a + self.epsilon)
new_p = p - lr * update
# Apply constraints.
if getattr(p, 'constraint', None) is not None:
new_p = p.constraint(new_p)
self.updates.append(K.update(p, new_p))
# update delta_accumulator
new_d_a = self.rho * d_a + (1 - self.rho) * K.square(update)
self.updates.append(K.update(d_a, new_d_a))
return self.updates
def reset_states(self):
if not self.stateful:
raise RuntimeError('Layer must be stateful.')
input_shape = self.input_spec[0].shape
output_shape = self._compute_output_shape(input_shape)
if not input_shape[0]:
raise ValueError('If a RNN is stateful, a complete '
'input_shape must be provided '
'(including batch size). '
'Got input shape: ' + str(input_shape))
if self.return_sequences:
out_row, out_col, out_filter = output_shape[2:]
else:
out_row, out_col, out_filter = output_shape[1:]
if hasattr(self, 'states'):
K.set_value(self.states[0],
np.zeros((input_shape[0], out_row, out_col, out_filter)))
K.set_value(self.states[1],
np.zeros((input_shape[0], out_row, out_col, out_filter)))
else:
self.states = [
K.zeros((input_shape[0], out_row, out_col, out_filter)),
K.zeros((input_shape[0], out_row, out_col, out_filter))
]
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
shapes = [K.int_shape(p) for p in params]
accumulators = [K.zeros(shape) for shape in shapes]
delta_accumulators = [K.zeros(shape) for shape in shapes]
self.weights = accumulators + delta_accumulators
self.updates = []
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * self.iterations))
self.updates.append(K.update_add(self.iterations, 1))
for p, g, a, d_a in zip(params, grads, accumulators,
delta_accumulators):
# update accumulator
new_a = self.rho * a + (1. - self.rho) * K.square(g)
self.updates.append(K.update(a, new_a))
# use the new accumulator and the *old* delta_accumulator
update = g * K.sqrt(d_a + self.epsilon) / K.sqrt(new_a +
self.epsilon)
new_p = p - lr * update
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
# update delta_accumulator
new_d_a = self.rho * d_a + (1 - self.rho) * K.square(update)
self.updates.append(K.update(d_a, new_d_a))
return self.updates
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
self.updates = [K.update_add(self.iterations, 1)]
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * self.iterations))
t = self.iterations + 1
lr_t = lr * (K.sqrt(1. - K.pow(self.beta_2, t)) /
(1. - K.pow(self.beta_1, t)))
shapes = [K.int_shape(p) for p in params]
ms = [K.zeros(shape) for shape in shapes]
vs = [K.zeros(shape) for shape in shapes]
self.weights = [self.iterations] + ms + vs
for p, g, m, v in zip(params, grads, ms, vs):
m_t = (self.beta_1 * m) + (1. - self.beta_1) * g
v_t = (self.beta_2 * v) + (1. - self.beta_2) * K.square(g)
p_t = p - lr_t * m_t / (K.sqrt(v_t) + self.epsilon)
self.updates.append(K.update(m, m_t))
self.updates.append(K.update(v, v_t))
new_p = p_t
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
self.updates = [K.update_add(self.iterations, 1)]
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * self.iterations))
t = self.iterations + 1
lr_t = lr * (K.sqrt(1. - K.pow(self.beta_2, t)) /
(1. - K.pow(self.beta_1, t)))
shapes = [K.int_shape(p) for p in params]
ms = [K.zeros(shape) for shape in shapes]
vs = [K.zeros(shape) for shape in shapes]
self.weights = [self.iterations] + ms + vs
for p, g, m, v in zip(params, grads, ms, vs):
m_t = (self.beta_1 * m) + (1. - self.beta_1) * g
v_t = (self.beta_2 * v) + (1. - self.beta_2) * K.square(g)
p_t = p - lr_t * m_t / (K.sqrt(v_t) + self.epsilon)
self.updates.append(K.update(m, m_t))
self.updates.append(K.update(v, v_t))
new_p = p_t
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def reset_states(self):
if not self.stateful:
raise RuntimeError('Layer must be stateful.')
input_shape = self.input_spec.shape
output_shape = self._compute_output_shape(input_shape)
if not input_shape[0]:
raise ValueError('If a RNN is stateful, a complete '
'input_shape must be provided '
'(including batch size). '
'Got input shape: ' + str(input_shape))
if self.return_sequences:
out_row, out_col, out_filter = output_shape[2:]
else:
out_row, out_col, out_filter = output_shape[1:]
if hasattr(self, 'states'):
K.set_value(self.states[0],
np.zeros((input_shape[0], out_row, out_col, out_filter)))
K.set_value(self.states[1],
np.zeros((input_shape[0], out_row, out_col, out_filter)))
else:
self.states = [
K.zeros((input_shape[0], out_row, out_col, out_filter)), K.zeros(
(input_shape[0], out_row, out_col, out_filter))
]
def get_updates(self, loss, params):
grads = self.get_gradients(loss, params)
self.updates = [K.update_add(self.iterations, 1)]
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * K.cast(self.iterations,
K.dtype(self.decay))))
t = K.cast(self.iterations, K.floatx()) + 1
lr_t = lr / (1. - K.pow(self.beta_1, t))
shapes = [K.int_shape(p) for p in params]
# zero init of 1st moment
ms = [K.zeros(shape) for shape in shapes]
# zero init of exponentially weighted infinity norm
us = [K.zeros(shape) for shape in shapes]
self.weights = [self.iterations] + ms + us
for p, g, m, u in zip(params, grads, ms, us):
m_t = (self.beta_1 * m) + (1. - self.beta_1) * g
u_t = K.maximum(self.beta_2 * u, K.abs(g))
p_t = p - lr_t * m_t / (u_t + self.epsilon)
self.updates.append(K.update(m, m_t))
self.updates.append(K.update(u, u_t))
new_p = p_t
# Apply constraints.
if getattr(p, 'constraint', None) is not None:
new_p = p.constraint(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
self.updates = []
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * self.iterations))
self.updates.append(K.update_add(self.iterations, 1))
# momentum
shapes = [K.int_shape(p) for p in params]
moments = [K.zeros(shape) for shape in shapes]
self.weights = [self.iterations] + moments
for p, g, m in zip(params, grads, moments):
v = self.momentum * m - lr * g # velocity
self.updates.append(K.update(m, v))
if self.nesterov:
new_p = p + self.momentum * v - lr * g
else:
new_p = p + v
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
self.updates = []
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * self.iterations))
self.updates.append(K.update_add(self.iterations, 1))
# momentum
shapes = [K.int_shape(p) for p in params]
moments = [K.zeros(shape) for shape in shapes]
self.weights = [self.iterations] + moments
for p, g, m in zip(params, grads, moments):
v = self.momentum * m - lr * g # velocity
self.updates.append(K.update(m, v))
if self.nesterov:
new_p = p + self.momentum * v - lr * g
else:
new_p = p + v
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, loss, params):
grads = self.get_gradients(loss, params)
accumulators = [
K.zeros(K.int_shape(p), dtype=K.dtype(p)) for p in params
]
self.weights = accumulators
self.updates = [K.update_add(self.iterations, 1)]
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * K.cast(self.iterations,
K.dtype(self.decay))))
for p, g, a in zip(params, grads, accumulators):
# update accumulator
new_a = self.rho * a + (1. - self.rho) * K.square(g)
self.updates.append(K.update(a, new_a))
new_p = p - lr * g / (K.sqrt(new_a) + self.epsilon)
# Apply constraints.
if getattr(p, 'constraint', None) is not None:
new_p = p.constraint(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, loss, params):
grads = self.get_gradients(loss, params)
self.updates = [K.update_add(self.iterations, 1)]
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * K.cast(self.iterations,
K.dtype(self.decay))))
# momentum
shapes = [K.int_shape(p) for p in params]
moments = [K.zeros(shape) for shape in shapes]
self.weights = [self.iterations] + moments
for p, g, m in zip(params, grads, moments):
v = self.momentum * m - lr * g # velocity
self.updates.append(K.update(m, v))
if self.nesterov:
new_p = p + self.momentum * v - lr * g
else:
new_p = p + v
# Apply constraints.
if getattr(p, 'constraint', None) is not None:
new_p = p.constraint(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
self.updates = [K.update_add(self.iterations, 1)]
t = self.iterations + 1
# Due to the recommendations in [2], i.e. warming momentum schedule
momentum_cache_t = self.beta_1 * (1. - 0.5 *
(K.pow(0.96, t * self.schedule_decay)))
momentum_cache_t_1 = self.beta_1 * (1. - 0.5 *
(K.pow(0.96,
(t + 1) * self.schedule_decay)))
m_schedule_new = self.m_schedule * momentum_cache_t
m_schedule_next = self.m_schedule * momentum_cache_t * momentum_cache_t_1
self.updates.append((self.m_schedule, m_schedule_new))
shapes = [K.int_shape(p) for p in params]
ms = [K.zeros(shape) for shape in shapes]
vs = [K.zeros(shape) for shape in shapes]
self.weights = [self.iterations] + ms + vs
for p, g, m, v in zip(params, grads, ms, vs):
# the following equations given in [1]
g_prime = g / (1. - m_schedule_new)
m_t = self.beta_1 * m + (1. - self.beta_1) * g
m_t_prime = m_t / (1. - m_schedule_next)
v_t = self.beta_2 * v + (1. - self.beta_2) * K.square(g)
v_t_prime = v_t / (1. - K.pow(self.beta_2, t))
m_t_bar = (
1. - momentum_cache_t) * g_prime + momentum_cache_t_1 * m_t_prime
self.updates.append(K.update(m, m_t))
self.updates.append(K.update(v, v_t))
p_t = p - self.lr * m_t_bar / (K.sqrt(v_t_prime) + self.epsilon)
new_p = p_t
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
self.updates = [K.update_add(self.iterations, 1)]
t = self.iterations + 1
# Due to the recommendations in [2], i.e. warming momentum schedule
momentum_cache_t = self.beta_1 * (1. - 0.5 *
(K.pow(0.96, t * self.schedule_decay)))
momentum_cache_t_1 = self.beta_1 * (1. - 0.5 *
(K.pow(0.96,
(t + 1) * self.schedule_decay)))
m_schedule_new = self.m_schedule * momentum_cache_t
m_schedule_next = self.m_schedule * momentum_cache_t * momentum_cache_t_1
self.updates.append((self.m_schedule, m_schedule_new))
shapes = [K.int_shape(p) for p in params]
ms = [K.zeros(shape) for shape in shapes]
vs = [K.zeros(shape) for shape in shapes]
self.weights = [self.iterations] + ms + vs
for p, g, m, v in zip(params, grads, ms, vs):
# the following equations given in [1]
g_prime = g / (1. - m_schedule_new)
m_t = self.beta_1 * m + (1. - self.beta_1) * g
m_t_prime = m_t / (1. - m_schedule_next)
v_t = self.beta_2 * v + (1. - self.beta_2) * K.square(g)
v_t_prime = v_t / (1. - K.pow(self.beta_2, t))
m_t_bar = (
1. - momentum_cache_t) * g_prime + momentum_cache_t_1 * m_t_prime
self.updates.append(K.update(m, m_t))
self.updates.append(K.update(v, v_t))
p_t = p - self.lr * m_t_bar / (K.sqrt(v_t_prime) + self.epsilon)
new_p = p_t
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_initial_states(self, inputs):
# (samples, timesteps, rows, cols, filters)
initial_state = K.zeros_like(inputs)
# (samples, rows, cols, filters)
initial_state = K.sum(initial_state, axis=1)
shape = list(self.kernel_shape)
shape[-1] = self.filters
initial_state = self.input_conv(
initial_state, K.zeros(tuple(shape)), padding=self.padding)
initial_states = [initial_state for _ in range(2)]
return initial_states
def get_initial_state(self, inputs):
# (samples, timesteps, rows, cols, filters)
initial_state = K.zeros_like(inputs)
# (samples, rows, cols, filters)
initial_state = K.sum(initial_state, axis=1)
shape = list(self.kernel_shape)
shape[-1] = self.filters
initial_state = self.input_conv(
initial_state, K.zeros(tuple(shape)), padding=self.padding)
initial_states = [initial_state for _ in range(2)]
return initial_states
def reset_states(self, states_value=None):
if not self.stateful:
raise AttributeError('Layer must be stateful.')
if not self.input_spec:
raise RuntimeError('Layer has never been called '
'and thus has no states.')
batch_size = self.input_spec.shape[0]
if not batch_size:
raise ValueError('If a RNN is stateful, it needs to know '
'its batch size. Specify the batch size '
'of your input tensors: \n'
'- If using a Sequential model, '
'specify the batch size by passing '
'a `batch_input_shape` '
'argument to your first layer.\n'
'- If using the functional API, specify '
'the time dimension by passing a '
'`batch_shape` argument to your Input layer.')
if states_value is not None:
if not isinstance(states_value, (list, tuple)):
states_value = [states_value]
if len(states_value) != len(self.states):
raise ValueError('The layer has ' + str(len(self.states)) +
' states, but the `states_value` '
'argument passed '
'only has ' + str(len(states_value)) + ' entries')
if self.states[0] is None:
self.states = [K.zeros((batch_size, self.units)) for _ in self.states]
if not states_value:
return
for i, state in enumerate(self.states):
if states_value:
value = states_value[i]
if value.shape != (batch_size, self.units):
raise ValueError('Expected state #' + str(i) + ' to have shape ' +
str((batch_size, self.units)) +
' but got array with shape ' + str(value.shape))
else:
value = np.zeros((batch_size, self.units))
K.set_value(state, value)
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
shapes = [K.int_shape(p) for p in params]
accumulators = [K.zeros(shape) for shape in shapes]
self.weights = accumulators
self.updates = []
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * self.iterations))
self.updates.append(K.update_add(self.iterations, 1))
for p, g, a in zip(params, grads, accumulators):
new_a = a + K.square(g) # update accumulator
self.updates.append(K.update(a, new_a))
new_p = p - lr * g / (K.sqrt(new_a) + self.epsilon)
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_updates(self, params, constraints, loss):
grads = self.get_gradients(loss, params)
shapes = [K.int_shape(p) for p in params]
accumulators = [K.zeros(shape) for shape in shapes]
self.weights = accumulators
self.updates = []
lr = self.lr
if self.initial_decay > 0:
lr *= (1. / (1. + self.decay * self.iterations))
self.updates.append(K.update_add(self.iterations, 1))
for p, g, a in zip(params, grads, accumulators):
new_a = a + K.square(g) # update accumulator
self.updates.append(K.update(a, new_a))
new_p = p - lr * g / (K.sqrt(new_a) + self.epsilon)
# apply constraints
if p in constraints:
c = constraints[p]
new_p = c(new_p)
self.updates.append(K.update(p, new_p))
return self.updates
def get_constants(self, inputs, training=None):
constants = []
if self.implementation == 0 and 0 < self.dropout < 1:
ones = K.zeros_like(inputs)
ones = K.sum(ones, axis=1)
ones += 1
def dropped_inputs():
return K.dropout(ones, self.dropout)
dp_mask = [
K.in_train_phase(dropped_inputs, ones, training=training)
for _ in range(4)
]
constants.append(dp_mask)
else:
constants.append([K.cast_to_floatx(1.) for _ in range(4)])
if 0 < self.recurrent_dropout < 1:
shape = list(self.kernel_shape)
shape[-1] = self.filters
ones = K.zeros_like(inputs)
ones = K.sum(ones, axis=1)
ones = self.input_conv(ones, K.zeros(shape), padding=self.padding)
ones += 1.
def dropped_inputs(): # pylint: disable=function-redefined
return K.dropout(ones, self.recurrent_dropout)
rec_dp_mask = [
K.in_train_phase(dropped_inputs, ones, training=training)
for _ in range(4)
]
constants.append(rec_dp_mask)
else:
constants.append([K.cast_to_floatx(1.) for _ in range(4)])
return constants
def get_constants(self, inputs, training=None):
constants = []
if self.implementation == 0 and 0 < self.dropout < 1:
ones = K.zeros_like(inputs)
ones = K.sum(ones, axis=1)
ones += 1
def dropped_inputs():
return K.dropout(ones, self.dropout)
dp_mask = [
K.in_train_phase(dropped_inputs, ones, training=training)
for _ in range(4)
]
constants.append(dp_mask)
else:
constants.append([K.cast_to_floatx(1.) for _ in range(4)])
if 0 < self.recurrent_dropout < 1:
shape = list(self.kernel_shape)
shape[-1] = self.filters
ones = K.zeros_like(inputs)
ones = K.sum(ones, axis=1)
ones = self.input_conv(ones, K.zeros(shape), padding=self.padding)
ones += 1.
def dropped_inputs(): # pylint: disable=function-redefined
return K.dropout(ones, self.recurrent_dropout)
rec_dp_mask = [
K.in_train_phase(dropped_inputs, ones, training=training)
for _ in range(4)
]
constants.append(rec_dp_mask)
else:
constants.append([K.cast_to_floatx(1.) for _ in range(4)])
return constants
def reset_states(self, states=None):
if not self.stateful:
raise AttributeError('Layer must be stateful.')
batch_size = self.input_spec[0].shape[0]
if not batch_size:
raise ValueError('If a RNN is stateful, it needs to know '
'its batch size. Specify the batch size '
'of your input tensors: \n'
'- If using a Sequential model, '
'specify the batch size by passing '
'a `batch_input_shape` '
'argument to your first layer.\n'
'- If using the functional API, specify '
'the time dimension by passing a '
'`batch_shape` argument to your Input layer.')
# initialize state if None
if self.states[0] is None:
self.states = [K.zeros((batch_size, self.units)) for _ in self.states]
elif states is None:
for state in self.states:
K.set_value(state, np.zeros((batch_size, self.units)))
else:
if not isinstance(states, (list, tuple)):
states = [states]
if len(states) != len(self.states):
raise ValueError('Layer ' + self.name + ' expects ' +
str(len(self.states)) + ' states, '
'but it received ' + str(len(states)) +
' state values. Input received: ' + str(states))
for index, (value, state) in enumerate(zip(states, self.states)):
if value.shape != (batch_size, self.units):
raise ValueError('State ' + str(index) +
' is incompatible with layer ' + self.name +
': expected shape=' + str((batch_size, self.units)) +
', found shape=' + str(value.shape))
K.set_value(state, value)
