108 lines
4.5 KiB
Python
108 lines
4.5 KiB
Python
# Copyright (C) 2018-2021 Intel Corporation
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
|
|
import math
|
|
|
|
import numpy as np
|
|
|
|
from extensions.ops.MatMul import MatMul
|
|
from extensions.ops.elementwise import Add, Mul
|
|
from extensions.ops.transpose import Transpose
|
|
from mo.front.common.partial_infer.utils import int64_array
|
|
from mo.front.common.replacement import FrontReplacementSubgraph
|
|
from mo.front.subgraph_matcher import SubgraphMatch
|
|
from mo.graph.graph import Graph, rename_nodes
|
|
from mo.ops.reshape import Reshape
|
|
|
|
|
|
class FullyConnectedDecomposer(FrontReplacementSubgraph):
|
|
"""
|
|
Decomposes FC operation:
|
|
1. Biases are added separately with the help of Add node
|
|
2. FC node itself is converted to MatMul
|
|
"""
|
|
enabled = True
|
|
|
|
def pattern(self):
|
|
return dict(
|
|
nodes=[('op', dict(kind='op', type='FullyConnected'))],
|
|
edges=[]
|
|
)
|
|
|
|
def replace_sub_graph(self, graph: Graph, match: [dict, SubgraphMatch]):
|
|
node = match['op']
|
|
name = node.soft_get('name', node.id)
|
|
|
|
# biases normalization
|
|
if 2 in node.in_ports() and not node.in_port(2).disconnected():
|
|
bias_node = Add(graph, {'name': name + '/Bias_'}).create_node()
|
|
node_name = node.name + '/WithoutBiases'
|
|
bias_node_name = node.name
|
|
rename_nodes([(node, node_name), (bias_node, bias_node_name)])
|
|
node.out_port(0).get_connection().set_source(bias_node.out_port(0))
|
|
node.in_port(2).get_connection().set_destination(bias_node.in_port(1))
|
|
node.out_port(0).connect(bias_node.in_port(0))
|
|
|
|
# weights normalization
|
|
assert node.has_valid('out-size')
|
|
out_size = node['out-size']
|
|
reshape_dim = int64_array([-1, out_size])
|
|
if node.has_and_set('transpose_weights'):
|
|
reshape_dim = int64_array([out_size, -1])
|
|
node.insert_op_on_input_port(in_port_idx=1, new_op_class=Reshape,
|
|
new_op_attrs={'name': name + '/weights_reshape'}, value=reshape_dim)
|
|
if node.has_and_set('transpose_weights'):
|
|
node.insert_op_on_input_port(in_port_idx=1, new_op_class=Transpose,
|
|
new_op_attrs={'name': name + '/weights_transpose'}, value=int64_array([1, 0]))
|
|
|
|
# input normalization for 4D Caffe and MxNet FullyConnected
|
|
if graph.graph['fw'] in ['caffe', 'mxnet']:
|
|
node.insert_op_on_input_port(in_port_idx=0, new_op_class=Reshape,
|
|
new_op_attrs={'name': name + '/flatten_fc_input'}, value=int64_array([0, -1]))
|
|
|
|
MatMul.update_node_stat(node, {})
|
|
|
|
|
|
class GemmDecomposer(FrontReplacementSubgraph):
|
|
"""
|
|
Decomposes Gemm operation:
|
|
1. Biases are added separately with the help of Add node
|
|
2. Multiplication by `alpha` and `beta` values are separated to Mul operations
|
|
3. Gemm operation itself is converted to MatMul
|
|
"""
|
|
enabled = True
|
|
|
|
def pattern(self):
|
|
return dict(
|
|
nodes=[('op', dict(kind='op', op='Gemm'))],
|
|
edges=[],
|
|
)
|
|
|
|
def replace_sub_graph(self, graph: Graph, match: [dict, SubgraphMatch]):
|
|
node = match['op']
|
|
name = node.soft_get('name', node.id)
|
|
|
|
# biases normalization
|
|
bias_node = Add(graph, {'name': name + '/Bias_', 'can_be_scaleshift': False}).create_node()
|
|
node_name = node.name + '/WithoutBiases'
|
|
bias_node_name = node.name
|
|
rename_nodes([(node, node_name), (bias_node, bias_node_name)])
|
|
node.out_port(0).get_connection().set_source(bias_node.out_port(0))
|
|
node.in_port(2).get_connection().set_destination(bias_node.in_port(1))
|
|
node.out_port(0).connect(bias_node.in_port(0))
|
|
|
|
if node.has_valid('alpha') and not math.isclose(node.alpha, 1):
|
|
bias_node.insert_op_on_input_port(in_port_idx=0, new_op_class=Mul, value=np.array(node.alpha),
|
|
new_op_attrs={'name': name + '/Alpha_', 'can_be_scaleshift': False})
|
|
del node['alpha']
|
|
|
|
if node.has_valid('beta') and not math.isclose(node.beta, 1):
|
|
bias_node.insert_op_on_input_port(in_port_idx=1, new_op_class=Mul, value=np.array(node.beta),
|
|
new_op_attrs={'name': name + '/Beta_', 'can_be_scaleshift': False})
|
|
del node['beta']
|
|
|
|
MatMul.update_node_stat(node, {
|
|
'transpose_a': node.has_and_set('transpose_a'),
|
|
'transpose_b': node.has_and_set('transpose_b'),
|
|
})
|