def __import_graph(cls, graph_path, session, out_nodes_names):
"""
:type graph_path: str
:type session: tensorflow.Session
:type out_nodes_names: list[str]
:rtype: tf.GraphDef
"""
if not os.path.exists(graph_path):
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_GRAPH_FILE_DOES_NOT_EXIST')(graph_path))
graph_path = os.path.abspath(graph_path)
if graph_path.endswith('.meta'):
checkpoint = graph_path.split('.meta')[0]
graph_def = cls.__import_from_meta_graph(graph_path, checkpoint,
out_nodes_names)
else:
graph_def = cls.__import_from_frozen_graph(graph_path)
if len(graph_def.node) == 0:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_NODES_NOT_FOUND_IN_GRAPH'))
with session.graph.as_default():
tf.import_graph_def(graph_def, name="")
return graph_def
def _convert_input_layers(self):
"""
:rtype: None
"""
for model_input in self._context.inputs:
input_operation = self._context.graph.get_operation_by_name(
model_input.name)
shape = self._graph_helper.get_op_output_shape(input_operation)
if None in shape:
message = code_to_message.get_message(
'ERROR_TF_UNABLE_TO_RESOLVE_GRAPH_INPUT_DIMS')
raise ConverterError(message(model_input.name))
if model_input.shape != shape:
message = code_to_message.get_message(
'ERROR_TF_UNEXPECTED_INPUT_SHAPE')
raise ConverterError(message(model_input.shape, shape))
self._logger.info(
code_to_message.get_progress_message(
'INFO_TF_BUILDING_INPUT_LAYER')(input_operation.name,
shape))
layer_name = str(input_operation.outputs[0].name)
descriptor = InputLayerDescriptor(layer_name, [input_operation])
self._input_descriptors.append(descriptor)
self._ops.remove(input_operation)
self._context.model.add_data_layer(descriptor.output_names[0],
shape, 'rgb', 'rgb',
model_input.type)
def resolve_layer(self, graph_matcher, graph_helper):
potential_descriptors = []
for sequence in self.sequences:
matches = graph_matcher.match_sequence(sequence)
for match in matches:
extract_glimpse = match['extract_glimpse']
size = match['size']
size_value = graph_helper.evaluate_tensor_output(size.outputs[0])
if size_value.size != 2:
raise ConverterError(
code_to_message.get_message('ERROR_TF_RESOLVE_EXTRACT_GLIMPSE_SIZE'))
offsets = match['offsets']
offsets_value = graph_helper.evaluate_tensor_output(offsets.outputs[0])
if len(offsets_value.shape) != 2 or offsets_value.shape[1] != 2:
raise ConverterError(
code_to_message.get_message('ERROR_TF_RESOLVE_EXTRACT_GLIMPSE_OFFSETS'))
output_op_nodes_names = [str(extract_glimpse.outputs[0].name)]
consumed_nodes = match.consumed_nodes
centered = bool(extract_glimpse.get_attr('centered'))
normalized = bool(extract_glimpse.get_attr('normalized'))
uniform_noise = bool(extract_glimpse.get_attr('uniform_noise'))
potential_descriptors.append(
ExtractGlimpseLayerResolver.Descriptor(str(extract_glimpse.name), consumed_nodes,
size_value[1], size_value[0],
offsets_value, centered,normalized,
uniform_noise,
output_names=output_op_nodes_names))
return potential_descriptors
def resolve_layer(self, graph_matcher, graph_helper):
descriptors = []
for sequence in self.sequences:
for match in graph_matcher.match_sequence(sequence):
pad_op = match['root']
input_op = match['input']
paddings_op = match['paddings']
paddings_tensor = graph_helper.evaluate_tensor_output(paddings_op.outputs[0])
paddings_shape = graph_helper.get_op_output_shape(paddings_op)
input_rank = len(graph_helper.get_op_output_shape(input_op))
if [input_rank, 2] != paddings_shape:
raise ConverterError(code_to_message.get_message(
'ERROR_TF_PAD_INVALUD_PADDINGS')(str([input_rank, 2]), str(paddings_shape)))
if 'const_values' in match:
const_values_op = match['const_values']
const_values = graph_helper.evaluate_tensor_output(const_values_op.outputs[0])
else:
const_values = 0.0
if not np.isscalar(const_values):
raise ConverterError(code_to_message.get_message('ERROR_TF_PAD_CONSTANT_NOT_SCALAR'))
consumed_nodes = match.consumed_nodes
pad_descriptor = PadLayerResolver.Descriptor(
str(pad_op.name), consumed_nodes, paddings_tensor,
snpe.modeltools.PADDING_CONSTANT, const_values,
output_names=[str(pad_op.outputs[0].name)])
descriptors.extend([pad_descriptor])
return descriptors
def resolve_layer(self, graph_matcher, graph_helper):
matches = graph_matcher.match_sequence(self.sequence)
if len(matches) == 0:
return []
descriptors = []
for match in matches:
variance_op = match['variance']
epsilon_op = match['epsilon']
if variance_op.type not in ['Identity', 'Const']:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_BATCHNORM_RESOLVE_VARIANCE'))
variance = graph_helper.evaluate_tensor_output(
variance_op.outputs[0])
if epsilon_op.type not in ['Identity', 'Const']:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_BATCHNORM_RESOLVE_EPSILON'))
epsilon = graph_helper.evaluate_tensor_output(
epsilon_op.outputs[0])
scale_op = match['scale']
if scale_op.type not in ['Identity', 'Const', 'Fill']:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_BATCHNORM_RESOLVE_SCALE'))
scale = graph_helper.evaluate_tensor_output(scale_op.outputs[0])
mean_op = match['mean']
if mean_op.type not in ['Identity', 'Const']:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_BATCHNORM_RESOLVE_MEAN'))
mean = graph_helper.evaluate_tensor_output(mean_op.outputs[0])
beta_op = match['beta']
if beta_op.type not in ['Identity', 'Const']:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_BATCHNORM_RESOLVE_BETA'))
beta = graph_helper.evaluate_tensor_output(beta_op.outputs[0])
output_op_nodes_names = [
str(match[node.identifier].outputs[0].name)
for node in self.sequence.output_nodes
]
descriptors.append(
BatchNormLayerResolver.Descriptor(
str(match['d'].name),
match.consumed_nodes,
bn_mul_op=match['d'],
mean=mean,
variance=variance,
epsilon=epsilon,
scale=scale,
beta=beta,
output_names=output_op_nodes_names))
return descriptors
def resolve_layer(self, graph_matcher, graph_helper):
matches = graph_matcher.match_sequence(self.sequence)
if len(matches) == 0:
return []
descriptors = []
for match in matches:
conv_trans_op = match['root']
_, weights_tensor, input_tensor = GraphHelper.get_op_input_tensors(
conv_trans_op, ('?', '?', '?'))
if weights_tensor.op.type not in ['Identity', 'Const']:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_DECONV_CANT_FIND_WEIGHTS_NODE'))
strides = conv_trans_op.get_attr('strides')
padding = conv_trans_op.get_attr('padding')
weights = graph_helper.evaluate_tensor_output(weights_tensor)
consumed_nodes = match.consumed_nodes
output_op_nodes_names = [
str(match[node.identifier].outputs[0].name)
for node in self.sequence.output_nodes
]
bias_op = None
try:
output_ops = graph_helper.get_op_outputs(conv_trans_op)
bias_op = GraphHelper.filter_single_op_by_type(
output_ops, 'BiasAdd')
_, biases = GraphHelper.get_op_input_tensors(
bias_op, ('?', '?'))
if biases.op.type not in ['Const', 'Identity']:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_DECONV_CANT_FIND_BIAS_NODE'))
biases = graph_helper.evaluate_tensor_output(biases)
consumed_nodes.append(bias_op)
output_op_nodes_names = [str(bias_op.outputs[0].name)]
except OperationNotFoundError:
biases = np.zeros(np.shape(weights)[-2], dtype=np.float32)
descriptors.append(
DeConvolutionOptimizedLayerResolver.Descriptor(
str(conv_trans_op.name),
consumed_nodes,
conv_trans_op,
bias_op,
weights,
strides,
padding,
biases,
input_tensor,
output_names=output_op_nodes_names))
return descriptors
def _create_layer_builder(cls, descriptor):
builder_class = layers.layer_builders.get(type(descriptor), None)
if builder_class is None:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_NO_INPUT_TO_CREATE_LAYER')(type(descriptor)))
return builder_class()
def resolve_layer(self, graph_matcher, graph_helper):
descriptors = []
for sequence in self.sequences:
for match in graph_matcher.match_sequence(sequence):
transpose_op = match['root']
input_op = match['input']
order_op = match['order']
order_tensor = graph_helper.evaluate_tensor_output(order_op.outputs[0])
input_shape = graph_helper.get_op_output_shape(input_op)
order_shape = graph_helper.get_op_output_shape(order_op)
input_rank = len(input_shape)
order_rank = len(order_shape)
try:
assert order_rank == 1
for d in range(input_rank):
assert d in order_tensor
except AssertionError:
raise ConverterError(code_to_message.get_message(
'ERROR_TF_PERMUTE_INVALID_ORDER_TENSOR')(str(order_tensor)))
consumed_nodes = match.consumed_nodes
permute_descriptor = PermuteLayerResolver.Descriptor(
str(transpose_op.name), consumed_nodes, order_tensor,
output_names=[str(transpose_op.outputs[0].name)])
descriptors.extend([permute_descriptor])
return descriptors
def resolve_layer(self, graph_matcher, graph_helper):
matches = graph_matcher.match_sequence(self.sequence)
potential_descriptors = []
for match in matches:
bn_op = match['root']
parameter_tensors = self._get_parameter_tensors(
graph_helper, bn_op)
if len(parameter_tensors) < 4:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_BATCHNORM_GLOBALNORMALIZATION_INPUT'))
epsilon = bn_op.get_attr('epsilon')
scale = parameter_tensors[0]
beta = parameter_tensors[1]
mean = parameter_tensors[2]
variance = parameter_tensors[3]
consumed_nodes = match.consumed_nodes
potential_descriptors.append(
FusedBatchNormNormLayerResolver.Descriptor(str(bn_op.name),
consumed_nodes,
bn_mul_op=bn_op,
mean=mean,
variance=variance,
epsilon=epsilon,
scale=scale,
beta=beta))
return potential_descriptors
def build_layer(self, converter_context, descriptor, input_descriptors, output_descriptors):
"""
:type input_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type output_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type converter_context: converters.tensorflow.converter.ConverterContext
:type descriptor: ConcatLayerResolver.Descriptor
:rtype: int
"""
input_names = self.get_input_names(converter_context, descriptor, input_descriptors)
if len(input_names) < 2:
raise ConverterError(code_to_message.get_message('ERROR_TF_ADD_N_NUM_OF_INPUTS')(descriptor.layer_name))
output_name = descriptor.output_names[0]
current_input_names = [input_names[0], input_names[1]]
current_output_name = descriptor.layer_name + '_unroll_1'
converter_context.model.add_elementwise_sum_layer(descriptor.layer_name + '_unroll_1',
[1.0 for _ in current_input_names],
current_input_names,
current_output_name)
for input_index in range(2, len(input_names) - 1):
current_input_names = [current_output_name, input_names[input_index]]
current_output_name = descriptor.layer_name + '_unroll_' + str(input_index)
converter_context.model.add_elementwise_sum_layer(descriptor.layer_name + '_unroll_' + str(input_index),
[1.0 for _ in current_input_names],
current_input_names,
current_output_name)
current_input_names = [current_output_name, input_names[-1]]
return converter_context.model.add_elementwise_sum_layer(descriptor.layer_name,
[1.0 for _ in current_input_names],
current_input_names,
output_name)
def _get_input_layers_output_tensors_for(self, operation):
"""
:type operation: tensorflow.Operation
:rtype: list[tensorflow.Tensor]
"""
descriptor = self._topology_resolver.descriptor_ops_map.get(
operation, None)
if descriptor is None:
raise ConverterError(
'Unable to find input layer for operation not in layer.')
output_tensors = []
input_descriptors = self._topology_resolver.get_input_layers_for(
descriptor)
input_descriptors_outputs = [
o for d in input_descriptors for o in d.child_ops
if d.is_output_op(o)
]
visited = set()
op_queue = [operation]
while len(op_queue) > 0:
next_op = op_queue.pop(0)
visited.add(next_op)
for input_tensor in next_op.inputs:
input_op = input_tensor.op
if input_op in input_descriptors_outputs:
output_tensors.append(input_tensor)
elif input_op not in visited:
op_queue.insert(0, input_op)
return uniques(output_tensors)
def resolve_layer(self, graph_matcher, graph_helper):
potential_descriptors = []
for sequence in self.sequences:
matches = graph_matcher.match_sequence(sequence)
for match in matches:
coefficients = match['alphas']
add_op = match['f']
if coefficients.type not in ['Identity', 'Const']:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_RESOLVE_PRELU_COEFF'))
output_op_nodes_names = [
str(match[node.identifier].outputs[0].name)
for node in sequence.output_nodes
]
consumed_nodes = match.consumed_nodes
potential_descriptors.append(
PReLuLayerResolver.Descriptor(
str(add_op.name),
consumed_nodes,
graph_helper.evaluate_tensor_output(
coefficients.outputs[0]),
output_names=output_op_nodes_names))
return potential_descriptors
def _get_input_layer_output_op_for(self, operation):
"""
:type operation: tensorflow.Operation
:rtype: tensorflow.Operation
"""
input_tensors = self._get_input_layers_output_tensors_for(operation)
ops = uniques([t.op for t in input_tensors])
if len(ops) == 0:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_INPUT_OPERATION_NOT_FOUND')(operation.name))
if len(ops) != 1:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_EXPECTED_SINGLE_OUTPUT_FROM_PREVIOUS_LAYER'))
return ops[0]
def __init__(self, nodes):
super(GraphSequence, self).__init__()
self._output_nodes = []
for node in nodes:
if node.identifier in self:
raise ConverterError('Node with id already defined {}'.format(
node.identifier))
self[node.identifier] = node
def get_biases(self, graph_helper, conv_op, bias_op):
_, biases_tensor = GraphHelper.get_op_input_tensors(
bias_op, ('?', '?'))
if biases_tensor.op.type not in ['Identity', 'Const']:
raise ConverterError(
code_to_message.get_message('ERROR_TF_CONV_RESOLVE_BIAS')(
conv_op.name))
biases = graph_helper.evaluate_tensor_output(biases_tensor)
return biases
def get_weights(self, graph_helper, conv_op):
_, weights_tensor = GraphHelper.get_op_input_tensors(
conv_op, ('?', '?'))
if weights_tensor.op.type not in ['Identity', 'Const', 'Split']:
raise ConverterError(
code_to_message.get_message('ERROR_TF_CONV_RESOLVE_WEIGHTS')(
conv_op.name))
weights = graph_helper.evaluate_tensor_output(weights_tensor)
return weights
def resolve_layer(self, graph_matcher, graph_helper):
descriptors = []
for match in graph_matcher.match_sequence(self.sequence):
strided_slice_op = match['root']
input_op = match['input']
if input_op.type == "Const":
continue
begin_op = match['begin']
end_op = match['end']
strides_op = match['strides']
begin_tensor = graph_helper.evaluate_tensor_output(
begin_op.outputs[0])
end_tensor = graph_helper.evaluate_tensor_output(end_op.outputs[0])
strides_tensor = graph_helper.evaluate_tensor_output(
strides_op.outputs[0])
input_tensor = graph_helper.evaluate_tensor_output(
input_op.outputs[0])
begin_shape = graph_helper.get_op_output_shape(begin_op)
end_shape = graph_helper.get_op_output_shape(end_op)
strides_shape = graph_helper.get_op_output_shape(strides_op)
input_shape = graph_helper.get_op_output_shape(input_op)
if begin_shape != end_shape or begin_shape != strides_shape:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_STRIDED_SLICE_SHAPE_MISMATCH'))
begin_mask = strided_slice_op.get_attr("begin_mask")
end_mask = strided_slice_op.get_attr("end_mask")
ellipsis_mask = strided_slice_op.get_attr("ellipsis_mask")
new_axis_mask = strided_slice_op.get_attr("new_axis_mask")
shrink_axis_mask = strided_slice_op.get_attr("shrink_axis_mask")
consumed_nodes = match.consumed_nodes
pad_descriptor = StridedSliceLayerResolver.Descriptor(
str(strided_slice_op.name),
consumed_nodes,
input_shape,
begin_tensor,
end_tensor,
strides_tensor,
begin_mask,
end_mask,
ellipsis_mask,
new_axis_mask,
shrink_axis_mask,
output_names=[str(strided_slice_op.outputs[0].name)])
descriptors.extend([pad_descriptor])
return descriptors
def resolve_layer(self, graph_matcher, graph_helper):
matches = graph_matcher.match_sequence(self.graph_sequence)
if len(matches) == 0:
return []
descriptor = []
for match in matches:
conv_op = match['conv_op']
strides = conv_op.get_attr(self.TF_ATTRIBUTE_STRIDES)
padding = conv_op.get_attr(self.TF_ATTRIBUTE_PADDING)
weights = self.get_weights(graph_helper, conv_op)
weights = np.transpose(weights, [0, 1, 3, 2])
consumed_nodes = match.consumed_nodes
output_op_nodes_names = [
str(match[node.identifier].outputs[0].name)
for node in self.graph_sequence.output_nodes
]
try:
batch_to_space_op = match['batch_to_space']
conv_output_ops = graph_helper.get_op_outputs(
batch_to_space_op)
bias_op = GraphHelper.filter_single_op_by_type(
conv_output_ops, 'BiasAdd')
biases = self.get_biases(graph_helper, conv_op, bias_op)
consumed_nodes.append(bias_op)
output_op_nodes_names = [str(bias_op.outputs[0].name)]
except OperationNotFoundError:
bias_op = None
biases = np.zeros(np.shape(weights)[-1], dtype=np.float32)
dilation_sizes = match['dilation_sizes']
dilation_sizes = graph_helper.evaluate_tensor_output(
dilation_sizes.outputs[0])
if np.shape(dilation_sizes) != (2, ):
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_CONV_RESOLVE_DILATION')(conv_op.name))
d = ConvolutionLayerResolver.Descriptor(
str(conv_op.name),
consumed_nodes,
conv_op,
bias_op,
strides,
padding,
weights,
biases,
output_names=output_op_nodes_names)
space_to_batch_op = match['space_to_batch']
d.groups = graph_helper.get_op_output_shape(space_to_batch_op)[-1]
d.dilationY = int(dilation_sizes[0])
d.dilationX = int(dilation_sizes[1])
d.input_ops = [space_to_batch_op]
descriptor.append(d)
return descriptor
def __import_from_meta_graph(cls, meta_graph_path, graph_path,
out_nodes_names):
"""
:type meta_graph_path: str
:type graph_path: str
:type out_nodes_names: list[str]
:rtype: tensorflow.GraphDef
"""
session = tf.Session(graph=tf.Graph())
with session.graph.as_default():
try:
saver = tf.train.import_meta_graph(meta_graph_path)
except AssertionError, e:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_CANNOT_IMPORT_GRAPH_FROM_META')(e.message))
if saver is None:
raise ConverterError(
code_to_message.get_message('ERROR_TF_GRAPH_META_EMPTY'))
saver.restore(session, graph_path)
def resolve_layer(self, graph_matcher, graph_helper):
matches = graph_matcher.match_sequence(self.sequence)
if len(matches) == 0:
return []
descriptors = []
for match in matches:
matmul_op = match['matmul_op']
weights_op = match['weights']
if weights_op.type not in ['Identity', 'Const']:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_MATMUL_RESOLVE_WEIGHTS')(matmul_op.name))
weights = graph_helper.evaluate_tensor_output(
weights_op.outputs[0])
bias_add_op = match['bias_op']
biases_op = match['biases']
if biases_op.type not in ['Identity', 'Const']:
# do we still need this check ?
raise ConverterError(
code_to_message.get_message('ERROR_TF_MATMUL_RESOLVE_BIAS')
(bias_add_op.name))
biases = graph_helper.evaluate_tensor_output(biases_op.outputs[0])
consumed_nodes = match.consumed_nodes
output_op_nodes_names = [
str(match[node.identifier].outputs[0].name)
for node in self.sequence.output_nodes
]
descriptors.append(
FullyConnectedLayerResolver.Descriptor(
str(matmul_op.name),
consumed_nodes,
matmul_op,
bias_add_op,
weights,
biases,
output_names=output_op_nodes_names))
return descriptors
def _assert_all_ops_consumed(self, descriptors, graph_ops):
graph_ops = self._filter_unconsumed_ops(descriptors, graph_ops)
def is_parameter_op(o):
return o.type in ['Const', 'Identity', 'Variable']
remaining_ops = [op for op in graph_ops if not is_parameter_op(op)]
for op in remaining_ops:
self._logger.warning(
code_to_message.get_warning_message(
'WARNING_TF_SCOPE_OP_NOT_CONSUMED')(op.name, op.type))
if len(remaining_ops) > 0:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_OPERATION_NOT_MAPPED_TO_LAYER'))
def build_layer(self, converter_context, descriptor, input_descriptors,
output_descriptors):
"""
:type input_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type output_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type converter_context: converters.tensorflow.converter.ConverterContext
:type descriptor: DeConvolutionLayerResolver.Descriptor
:rtype: int
"""
input_dims = converter_context.graph_helper.get_op_output_shape(
descriptor.input_tensor.op)
if descriptor.bias_op:
output_dims = converter_context.graph_helper.get_op_output_shape(
descriptor.bias_op)
else:
output_dims = converter_context.graph_helper.get_op_output_shape(
descriptor.deconv_op)
pad_y, pad_x, padding_strategy = ConvolutionLayerBuilder.calculate_padding_size(
input_size=output_dims[-3:-1],
output_size=input_dims[-3:-1],
strides=descriptor.strides[1:3],
padding=descriptor.padding,
filter_dims=descriptor.weights.shape,
dilation=[1, 1])
if pad_y != pad_x:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_DECONV_NO_SUPPORT_RECT_PADDING'))
weights = np.transpose(descriptor.weights, (0, 1, 3, 2)).copy()
input_names = self.get_input_name(converter_context, descriptor,
input_descriptors)
return converter_context.model.add_deconvolution_layer(
name=descriptor.layer_name,
weights=weights,
bias=descriptor.biases,
stride=descriptor.strides[1],
padding_size_strategy=padding_strategy,
padding=pad_y,
input_name=input_names,
output_name=descriptor.output_names[0],
output_width=output_dims[-2],
output_height=output_dims[-3],
groups=1)
def build_layer(self, converter_context, descriptor, input_descriptors,
output_descriptors):
"""
:type input_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type output_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type converter_context: converters.tensorflow.converter.ConverterContext
:type descriptor: ConcatLayerResolver.Descriptor
:rtype: int
"""
if len(input_descriptors) < 2:
raise ConverterError(
code_to_message.get_message('ERROR_TF_CONCAT_INPUT'))
input_names = self.get_input_names(converter_context, descriptor,
input_descriptors)
return converter_context.model.add_concatenation_layer(
descriptor.layer_name, input_names, descriptor.output_names[0],
descriptor.axis)
def resolve_layer(self, graph_matcher, graph_helper):
descriptors = []
for sequence in self.sequences:
for match in graph_matcher.match_sequence(sequence):
pow_op = match['pow']
const_values_op = match['const']
const_values = graph_helper.evaluate_tensor_output(const_values_op.outputs[0])
# only scalar power op is supported
if not np.isscalar(const_values):
raise ConverterError(code_to_message.get_message('ERROR_TF_POW_CONSTANT_NOT_SCALAR'))
consumed_nodes = match.consumed_nodes
pow_descriptor = PowLayerResolver.Descriptor(
str(pow_op.name), consumed_nodes, 1, 0, const_values,
output_names=[str(pow_op.outputs[0].name)])
descriptors.extend([pow_descriptor])
return descriptors
def resolve_layer(self, graph_matcher, graph_helper):
descriptors = []
for match in graph_matcher.match_sequence(self.sequence):
argmax_op = match['root']
input_op = match['input']
axis_op = match['axis']
input_shape = graph_helper.get_op_output_shape(input_op)
input_rank = len(input_shape)
axis = int(graph_helper.evaluate_tensor_output(axis_op.outputs[0]))
if axis < 0:
axis += input_rank
if axis < 0 or axis >= input_rank:
raise ConverterError(code_to_message.get_message('ERROR_TF_ARGMAX_INVALID_AXIS')(axis, input_rank))
consumed_nodes = match.consumed_nodes
argmax_descriptor = ArgMaxLayerResolver.Descriptor(
str(argmax_op.name), consumed_nodes, axis,
output_names=[str(argmax_op.outputs[0].name)])
descriptors.extend([argmax_descriptor])
return descriptors
def build_layer(self, converter_context, descriptor, input_descriptors,
output_descriptors):
"""
:type input_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type output_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type converter_context: converters.tensorflow.converter.ConverterContext
:type descriptor: LstmLayerResolver.UnrolledTimeStepDescriptor
:rtype: int
"""
if isinstance(descriptor, LstmLayerResolver.StateDescriptor):
return
input_descriptors = [
d for d in input_descriptors
if not isinstance(d, LstmLayerResolver.StateDescriptor)
]
if len(input_descriptors) not in [1, 3]:
raise ConverterError('LSTM layer requires 1 or 3 inputs')
input_shape = converter_context.graph_helper.get_op_output_shape(
descriptor.cell_input_concat_op.inputs[0].op)
state_shape = converter_context.graph_helper.get_op_output_shape(
descriptor.cell_input_concat_op.inputs[1].op)
gates_weights, input_weights = self._resolve_weights(
descriptor, converter_context.graph_helper, state_shape)
gates_biases = self._resolve_biases(descriptor,
converter_context.graph_helper)
def is_cell_input_descriptor(d):
output_shape = []
output_ops = [op for op in d.child_ops if d.is_output_op(op)]
if len(output_ops) > 0:
output_shape = converter_context.graph_helper.get_op_output_shape(
output_ops[0])
return len(output_shape
) == 2 and output_shape[1] == descriptor.time_steps()
cell_input_descriptors = filter(is_cell_input_descriptor,
input_descriptors)
cell_state_descriptors = [
d for d in input_descriptors if d not in cell_input_descriptors
]
user_initial_state = len(cell_state_descriptors) == 2
is_stacked_above_cell = self.is_stacked_cell(input_descriptors)
if not is_stacked_above_cell:
if len(cell_input_descriptors) != 1:
raise ConverterError(
'Unable to resolve LSTM input layer name.')
cell_input_name = cell_input_descriptors[0].output_names[0]
input_layer_name = self._add_reshape_to_restore_time_dimension(
converter_context, descriptor, cell_input_name, input_shape)
else:
input_layer_name = input_descriptors[0].output_names[0]
is_stacked_below_cell = self.is_stacked_cell(output_descriptors)
descriptor.set_is_stacked_cell(is_stacked_below_cell)
output_names = [descriptor.stacked_cell_output_name]
if user_initial_state or (not is_stacked_below_cell
and len(output_descriptors) > 0):
output_names.append('{}_state'.format(descriptor.output_names[0]))
output_names.append(descriptor.output_names[0])
h_0_input_name = cell_state_descriptors[0].output_names[
0] if user_initial_state else ''
c_0_input_name = cell_state_descriptors[1].output_names[
0] if user_initial_state else ''
return converter_context.model.add_lstm_layer(
name=descriptor.output_names[0],
w_xc=input_weights,
b_c=gates_biases,
w_hc=gates_weights,
w_xc_static=None,
backward=False,
reset_state_at_time_step_0=not user_initial_state,
input_name=input_layer_name,
sequence_continuation_input_name='',
x_static_input_name='',
c_0_input_name=c_0_input_name,
h_0_input_name=h_0_input_name,
output_names=output_names)
def resolve_layer(self, graph_matcher, graph_helper):
matches = graph_matcher.match_sequence(self.graph_sequence)
if len(matches) == 0:
return []
descriptors = []
for match in matches:
conv_op = match['conv_op']
strides = conv_op.get_attr(self.TF_ATTRIBUTE_STRIDES)
padding = conv_op.get_attr(self.TF_ATTRIBUTE_PADDING)
weights = self.get_weights(graph_helper, conv_op)
weights = np.transpose(weights, [0, 1, 3, 2])
consumed_nodes = match.consumed_nodes
output_op_nodes_names = [str(match[node.identifier].outputs[0].name) for node in
self.graph_sequence.output_nodes]
try:
batch_to_space_op = match['batch_to_space']
conv_output_ops = graph_helper.get_op_outputs(batch_to_space_op)
bias_op = GraphHelper.filter_single_op_by_type(conv_output_ops, 'BiasAdd')
biases = self.get_biases(graph_helper, conv_op, bias_op)
consumed_nodes.append(bias_op)
output_op_nodes_names = [str(bias_op.outputs[0].name)]
except OperationNotFoundError:
bias_op = None
biases = np.zeros(np.shape(weights)[-1], dtype=np.float32)
dilation_sizes = match['dilation_sizes']
dilation_sizes = graph_helper.evaluate_tensor_output(dilation_sizes.outputs[0])
if np.shape(dilation_sizes) != (2,):
raise ConverterError(code_to_message.get_message('ERROR_TF_CONV_RESOLVE_DILATION')(conv_op.name))
space_to_batch_op = match['space_to_batch']
paddings_op = match['paddings']
paddings_tensor = graph_helper.evaluate_tensor_output(paddings_op.outputs[0])
input_op = conv_op
batch_to_space_op = match['batch_to_space']
crop_op = match['crops']
crops_tensor = graph_helper.evaluate_tensor_output(crop_op.outputs[0])
output_names = [str(conv_op.outputs[0].name)]
if paddings_tensor.any() and not np.array_equal(paddings_tensor, crops_tensor):
paddings_tensor = np.pad(paddings_tensor, ((1, 1), (0, 0)), 'constant')
pad_descriptor = PadLayerResolver.Descriptor(
str(space_to_batch_op.name),
[match['space_to_batch'], match['dilation_sizes'], match['paddings']],
paddings_tensor,
snpe.modeltools.PADDING_CONSTANT,
0.0,
output_names=[str(space_to_batch_op.outputs[0].name)])
descriptors.append(pad_descriptor)
else:
consumed_nodes.extend([space_to_batch_op, paddings_op, match['dilation_sizes']])
input_op = space_to_batch_op
if crops_tensor.any() and not np.array_equal(paddings_tensor, crops_tensor):
crops_tensor = np.pad(crops_tensor, ((1, 1), (0, 0)), 'constant')
offsets = crops_tensor[:, 0]
size = np.array(graph_helper.get_op_output_shape(match['batch_to_space']), dtype=np.int32)
crop_descriptor = CropLayerResolver.Descriptor(
str(match['batch_to_space'].name),
[match['batch_to_space'], match['block_shape_out'], match['crops']],
offsets,
size,
output_names=[str(match['batch_to_space'].outputs[0].name)])
descriptors.append(crop_descriptor)
else:
consumed_nodes.extend([batch_to_space_op, crop_op, match['block_shape_out']])
output_names = output_op_nodes_names
d = ConvolutionLayerResolver.Descriptor(str(conv_op.name), consumed_nodes,
conv_op, bias_op, strides, padding, weights,
biases,
output_names=output_names)
d.groups = graph_helper.get_op_output_shape(space_to_batch_op)[-1]
d.dilationY = int(dilation_sizes[0])
d.dilationX = int(dilation_sizes[1])
d.input_ops = [input_op]
descriptors.append(d)
return descriptors
def _broadcast_tensor(self, tensor, shape):
raise ConverterError('ElementWise resolver must implement broadcast method.')
def resolve_layer(self, graph_matcher, graph_helper):
raise ConverterError('Constant layers are resolved by other resolvers!')
def build_layer(self, converter_context, descriptor, input_descriptors,
output_descriptors):
"""
:type input_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type output_descriptors: [converters.tensorflow.common.LayerDescriptor]
:type converter_context: converters.tensorflow.converter.ConverterContext
:type descriptor: StridedSliceLayerResolver.Descriptor
:rtype: int
"""
input_name = self.get_input_name(converter_context, descriptor,
input_descriptors)
output_name = descriptor.output_names[0]
if descriptor.ellipsis_mask != 0 or descriptor.new_axis_mask != 0:
raise ConverterError(
code_to_message.get_message(
'ERROR_TF_STRIDED_SLICE_UNSUPPORTED_MASKS'))
input_rank = len(descriptor.input_shape)
strides_rank = descriptor.strides.shape[0]
# Extend to match input rank
begin = np.append(descriptor.begin,
np.zeros(input_rank - strides_rank,
dtype=np.int32)).tolist()
strides = np.append(descriptor.strides,
np.ones(input_rank - strides_rank,
dtype=np.int32)).tolist()
end = np.append(descriptor.end,
descriptor.input_shape[strides_rank:]).astype(
np.int32).tolist()
# Apply the binary masks
for i in range(len(strides)):
begin_mask_bit = self.get_bit(descriptor.begin_mask, i)
end_mask_bit = self.get_bit(descriptor.end_mask, i)
shrink_mask_bit = self.get_bit(descriptor.shrink_axis_mask, i)
# Convert negative indices
if begin[i] < 0:
begin[i] += descriptor.input_shape[i]
if end[i] < 0:
end[i] += descriptor.input_shape[i]
# Apply mask bits
if strides[i] > 0:
if begin_mask_bit:
begin[i] = 0
if end_mask_bit:
end[i] = descriptor.input_shape[i]
else:
if begin_mask_bit:
begin[i] = descriptor.input_shape[i] - 1
if end_mask_bit:
end[i] = -1
# Apply shrink_axis_mask
if shrink_mask_bit:
strides[i] = 1
end[i] = begin[i] + strides[i]
return converter_context.model.add_strided_slice_layer(
name=descriptor.layer_name,
input_name=input_name,
output_name=output_name,
begin=begin,
end=end,
strides=strides,
shrink_axis_mask=descriptor.shrink_axis_mask)
