def create_variables(self, input_spaces, action_space=None):
in_space = input_spaces["inputs[0]"]
assert in_space.rank > 0, "ERROR: Must have input Space ({}) with rank larger 0!".format(
in_space)
# Create weights matrix and (maybe) biases vector.
weights_shape = (in_space.shape[0], self.units)
self.weights_init = Initializer.from_spec(
shape=weights_shape, specification=self.weights_spec)
biases_shape = (self.units, )
self.biases_init = Initializer.from_spec(
shape=biases_shape, specification=self.biases_spec)
# Wrapper for backend.
if get_backend() == "tf":
self.layer = tf.layers.Dense(
units=self.units,
activation=get_activation_function(self.activation,
*self.activation_params),
kernel_initializer=self.weights_init.initializer,
use_bias=(self.biases_spec is not False),
bias_initializer=(self.biases_init.initializer
or tf.zeros_initializer()),
trainable=(False if self.trainable is False else True),
_reuse=tf.AUTO_REUSE)
# Now build the layer so that its variables get created.
self.layer.build(in_space.get_shape(with_batch_rank=True))
# Register the generated variables with our registry.
self.register_variables(*self.layer.variables)
elif get_backend() == "pytorch":
# N.b. activation must be added as a separate 'layer' when assembling a network.
# In features is the num of input channels.
apply_bias = (self.biases_spec is not False)
in_features = in_space.shape[1] if in_space.shape[
0] == 1 else in_space.shape[0]
# print("name = {}, ndim = {}, in space.shape = {}, in_features = {}, units = {}".format(
# self.name, ndim, in_space.shape, in_features, self.units))
self.layer = nn.Linear(
# In case there is a batch dim here due to missing preprocessing.
in_features=in_features,
out_features=self.units,
bias=apply_bias)
# Apply weight initializer
if self.weights_init.initializer is not None:
# Must be a callable in PyTorch
self.weights_init.initializer(self.layer.weight)
if apply_bias:
if self.biases_spec is not None and self.biases_init.initializer is not None:
self.biases_init.initializer(self.layer.bias)
else:
# Fill with zeros.
self.layer.bias.data.fill_(0)
if self.activation is not None:
# Activation function will be used in apply.
self.activation_fn = get_activation_function(
self.activation, *self.activation_params)
# Use unique scope as name.
self.register_variables(
PyTorchVariable(name=self.global_scope, ref=self.layer))
def create_variables(self, input_spaces, action_space=None):
self.in_space = input_spaces["inputs"]
# Create one weight matrix: [input nodes + internal state nodes, 4 (4 internal layers) * internal state nodes]
# weights_shape = (in_space.shape[0] + self.units, 4 * self.units) # [0]=one past batch rank
# self.weights_init = Initializer.from_spec(shape=weights_shape, specification=self.weights_spec)
# Wrapper for backend.
if get_backend() == "tf":
self.lstm = tf.contrib.rnn.LSTMBlockCell( #tf.nn.rnn_cell.LSTMCell(
num_units=self.units,
use_peephole=self.use_peepholes,
cell_clip=self.cell_clip,
forget_bias=self.forget_bias,
name="lstm-cell",
reuse=tf.AUTO_REUSE
# TODO: self.trainable needs to be recognized somewhere here.
# These are all not supported yet for LSTMBlockCell (only for the slower LSTMCell)
# initializer=self.weights_init.initializer,
# activation=get_activation_function(self.activation, *self.activation_params),
# dtype=self.dtype,
)
# Now build the layer so that its variables get created.
in_space_without_time_rank = list(
self.in_space.get_shape(with_batch_rank=True))
self.lstm.build(tf.TensorShape(in_space_without_time_rank))
# Register the generated variables with our registry.
self.register_variables(*self.lstm.variables)
elif get_backend() == "pytorch":
self.lstm = nn.LSTM(self.in_space, self.units)
self.hidden_state = (torch.zeros(1, 1, self.units),
torch.zeros(1, 1, self.units))
self.register_variables(
PyTorchVariable(name=self.global_scope, ref=self.lstm))
def create_variables(self, input_spaces, action_space=None):
in_space = input_spaces["inputs[0]"]
# Create kernel and biases initializers.
self.kernel_init = Initializer.from_spec(
shape=self.kernel_size, specification=self.kernel_spec)
self.biases_init = Initializer.from_spec(
shape=self.kernel_size, specification=self.biases_spec)
# Wrapper for backend.
if get_backend() == "tf":
self.layer = tf.layers.Conv2D(
filters=self.filters,
kernel_size=self.kernel_size,
strides=self.strides,
padding=self.padding,
data_format=self.data_format,
activation=get_activation_function(self.activation,
*self.activation_params),
use_bias=(self.biases_spec is not False),
kernel_initializer=self.kernel_init.initializer,
bias_initializer=(self.biases_init.initializer
or tf.zeros_initializer()),
trainable=(False if self.trainable is False else True),
_reuse=tf.AUTO_REUSE)
# Now build the layer so that its variables get created.
self.layer.build(in_space.get_shape(with_batch_rank=True))
# Register the generated variables with our registry.
self.register_variables(*self.layer.variables)
elif get_backend() == "pytorch":
shape = in_space.shape
num_channels = get_input_channels(shape)
apply_bias = (self.biases_spec is not False)
# print("Defining conv2d layer with shape = {} and channels {}".format(
# shape, num_channels
# ))
if self.padding == "same":
# N.b. there is no 'same' or 'valid' padding for PyTorch so need custom layer.
self.layer = SamePaddedConv2d(
in_channels=num_channels,
out_channels=self.filters,
# Only support square kernels.
kernel_size=self.kernel_size[0],
stride=self.strides,
bias=apply_bias)
else:
self.layer = nn.Conv2d(in_channels=num_channels,
out_channels=self.filters,
kernel_size=self.kernel_size,
stride=self.strides,
padding=0,
bias=apply_bias)
# Apply weight initializer
if self.kernel_init.initializer is not None:
# Must be a callable in PyTorch
self.kernel_init.initializer(self.layer.weight)
if apply_bias:
if self.biases_spec is not None and self.biases_init.initializer is not None:
self.biases_init.initializer(self.layer.bias)
else:
# Fill with zeros.
self.layer.bias.data.fill_(0)
if self.activation is not None:
# Activation function will be used in `call`.
self.activation_fn = get_activation_function(
self.activation, *self.activation_params)
self.register_variables(
PyTorchVariable(name=self.global_scope, ref=self.layer))