72 lines
2.8 KiB
Python
72 lines
2.8 KiB
Python
# Copyright (C) 2018-2021 Intel Corporation
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
|
|
from mo.front.common.partial_infer.utils import convert_deconv_tf_padding_to_str, int64_array
|
|
from mo.front.extractor import FrontExtractorOp
|
|
from mo.front.tf.extractors.utils import tf_data_format_spatial, tf_data_format_channel, tf_data_format_batch, \
|
|
tf_int_list
|
|
from mo.ops.deconvolution import Deconvolution
|
|
from mo.ops.op import PermuteAttrs
|
|
|
|
|
|
class Conv2DBackpropInputFrontExtractor(FrontExtractorOp):
|
|
op = 'Conv2DBackpropInput'
|
|
enabled = True
|
|
|
|
@classmethod
|
|
def extract(cls, node):
|
|
attrs = tf_create_attrs(node, 3, 2)
|
|
attrs.update({'op': cls.op,
|
|
'get_weights_permute': PermuteAttrs.Permutation(perm=int64_array([3, 2, 0, 1]),
|
|
inv=int64_array([2, 3, 1, 0])),
|
|
'swap_0_and_2_inputs': True,
|
|
'shape_input': True,
|
|
})
|
|
|
|
# update the attributes of the node
|
|
Deconvolution.update_node_stat(node, attrs)
|
|
return cls.enabled
|
|
|
|
|
|
class Conv3DBackpropInputV2InputFrontExtractor(FrontExtractorOp):
|
|
op = 'Conv3DBackpropInputV2'
|
|
enabled = True
|
|
|
|
@classmethod
|
|
def extract(cls, node):
|
|
attrs = tf_create_attrs(node, 4, 3)
|
|
attrs.update({'op': cls.op,
|
|
'get_weights_permute': PermuteAttrs.Permutation(perm=int64_array([4, 3, 0, 1, 2]),
|
|
inv=int64_array([2, 3, 4, 1, 0])),
|
|
'swap_0_and_2_inputs': True,
|
|
'shape_input': True,
|
|
})
|
|
|
|
# update the attributes of the node
|
|
Deconvolution.update_node_stat(node, attrs)
|
|
return cls.enabled
|
|
|
|
|
|
def tf_create_attrs(node, input_feature_channel, output_feature_channel):
|
|
data_format = node.pb.attr["data_format"]
|
|
|
|
return {
|
|
'auto_pad': convert_deconv_tf_padding_to_str(node.pb.attr['padding'].s.decode()),
|
|
'bias_addable': True,
|
|
'bias_term': False,
|
|
'spatial_dims': tf_data_format_spatial(data_format),
|
|
'channel_dims': tf_data_format_channel(data_format),
|
|
'batch_dims': tf_data_format_batch(data_format),
|
|
'pad': None, # will be inferred when input shape is known
|
|
'pad_spatial_shape': None,
|
|
'output_spatial_shape': None,
|
|
'output_shape': None,
|
|
'output': None,
|
|
'stride': tf_int_list(node.pb.attr["strides"].list),
|
|
'type': None, # don't set type until we are sure it is really translated to correct IR; see infer function
|
|
'group': None,
|
|
'layout': data_format.s.decode(),
|
|
'input_feature_channel': input_feature_channel,
|
|
'output_feature_channel': output_feature_channel,
|
|
}
|