1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
|
// Copyright Verizon Media. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
#include <vespa/eval/eval/simple_value.h>
#include <vespa/eval/eval/value_codec.h>
#include <vespa/eval/eval/simple_tensor.h>
#include <vespa/eval/eval/simple_tensor_engine.h>
#include <vespa/eval/eval/value_codec.h>
#include <vespa/eval/instruction/generic_concat.h>
#include <vespa/eval/eval/interpreted_function.h>
#include <vespa/eval/eval/test/tensor_model.hpp>
#include <vespa/vespalib/util/stringfmt.h>
#include <vespa/vespalib/gtest/gtest.h>
using namespace vespalib;
using namespace vespalib::eval;
using namespace vespalib::eval::instruction;
using namespace vespalib::eval::test;
using vespalib::make_string_short::fmt;
std::vector<Layout> concat_layouts = {
{}, {},
{}, {y(5)},
float_cells({y(5)}), {},
{}, float_cells({y(5)}),
{y(5)}, {},
{y(2)}, {y(3)},
{y(2)}, {x(3)},
{x(2)}, {z(3)},
{x(2),y(3)}, {x(2),y(3)},
{x(2),y(3)}, {x(2),y(4)},
{y(3),z(5)}, {y(3),z(5)},
{y(3),z(5)}, {y(4),z(5)},
{x(2),y(3),z(5)}, {x(2),y(3),z(5)},
{x(2),y(3),z(5)}, {x(2),y(4),z(5)},
{x(2),y(3),z({"a","b"})}, {x(2),y(3),z({"b","c"})},
{x(2),y(3),z({"a","b"})}, {x(2),y(4),z({"b","c"})},
{y(5)}, {y(2),x(5)},
{x(3)}, {y(2),z(3)},
{y(2)}, {y(3),x(5),z(2)},
{y(2),x(5),z(2)}, {y(3),x(5),z(2)},
{y(3),x(5)}, {x(5),z(7)},
float_cells({y(3),x(5)}), {x(5),z(7)},
float_cells({y(3),x(5)}), {},
{y(3),x(5)}, float_cells({x(5),z(7)}),
float_cells({y(3),x(5)}), float_cells({x(5),z(7)}),
{x({"a","b","c"})}, {x({"a","b","c"})},
{x({"a","b","c"})}, {x({"a","b"})},
{x({"a","b","c"})}, {x({"b","c","d"})},
float_cells({x({"a","b","c"})}), {x({"b","c","d"})},
{x({"a","b","c"})}, float_cells({x({"b","c","d"})}),
float_cells({x({"a","b","c"})}), float_cells({z({"foo","bar","baz"})}),
{x({"a","b","c"})}, {x({"a","b","c"}),z({"foo","bar","baz"})},
{x({"a","b"}),z({"foo","bar","baz"})}, {x({"a","b","c"}),z({"foo","bar"})},
{y(2),x({"a","b","c"})}, {y(3),x({"b","c","d"})},
{y(2),x({"a","b"})}, {y(3),z({"c","d"})}
};
TensorSpec perform_simpletensor_concat(const TensorSpec &a, const TensorSpec &b, const std::string &dimension) {
auto lhs = SimpleTensor::create(a);
auto rhs = SimpleTensor::create(b);
auto out = SimpleTensor::concat(*lhs, *rhs, dimension);
return SimpleTensorEngine::ref().to_spec(*out);
}
bool concat_address(const TensorSpec::Address &me, const TensorSpec::Address &other,
const std::string &concat_dim, size_t my_offset,
TensorSpec::Address &my_out, TensorSpec::Address &other_out)
{
my_out.insert_or_assign(concat_dim, my_offset);
for (const auto &my_dim: me) {
const auto & name = my_dim.first;
const auto & label = my_dim.second;
if (name == concat_dim) {
my_out.insert_or_assign(name, label.index + my_offset);
} else {
auto pos = other.find(name);
if ((pos == other.end()) || (pos->second == label)) {
my_out.insert_or_assign(name, label);
other_out.insert_or_assign(name, label);
} else {
return false;
}
}
}
return true;
}
bool concat_addresses(const TensorSpec::Address &a, const TensorSpec::Address &b,
const std::string &concat_dim, size_t b_offset,
TensorSpec::Address &a_out, TensorSpec::Address &b_out)
{
return concat_address(a, b, concat_dim, 0, a_out, b_out) &&
concat_address(b, a, concat_dim, b_offset, b_out, a_out);
}
TensorSpec reference_concat(const TensorSpec &a, const TensorSpec &b, const std::string &concat_dim) {
ValueType a_type = ValueType::from_spec(a.type());
ValueType b_type = ValueType::from_spec(b.type());
ValueType res_type = ValueType::concat(a_type, b_type, concat_dim);
EXPECT_FALSE(res_type.is_error());
size_t b_offset = 1;
size_t concat_dim_index = a_type.dimension_index(concat_dim);
if (concat_dim_index != ValueType::Dimension::npos) {
const auto &dim = a_type.dimensions()[concat_dim_index];
EXPECT_TRUE(dim.is_indexed());
b_offset = dim.size;
}
TensorSpec result(res_type.to_spec());
for (const auto &cell_a: a.cells()) {
for (const auto &cell_b: b.cells()) {
TensorSpec::Address addr_a;
TensorSpec::Address addr_b;
if (concat_addresses(cell_a.first, cell_b.first, concat_dim, b_offset, addr_a, addr_b)) {
result.add(addr_a, cell_a.second);
result.add(addr_b, cell_b.second);
}
}
}
return result;
}
TensorSpec perform_generic_concat(const TensorSpec &a, const TensorSpec &b, const std::string &concat_dim) {
Stash stash;
const auto &factory = SimpleValueBuilderFactory::get();
auto lhs = value_from_spec(a, factory);
auto rhs = value_from_spec(b, factory);
auto my_op = GenericConcat::make_instruction(lhs->type(), rhs->type(), concat_dim, factory, stash);
InterpretedFunction::EvalSingle single(factory, my_op);
return spec_from_value(single.eval(std::vector<Value::CREF>({*lhs,*rhs})));
}
TEST(GenericConcatTest, generic_reference_concat_works) {
ASSERT_TRUE((concat_layouts.size() % 2) == 0);
for (size_t i = 0; i < concat_layouts.size(); i += 2) {
const TensorSpec lhs = spec(concat_layouts[i], N());
const TensorSpec rhs = spec(concat_layouts[i + 1], Div16(N()));
SCOPED_TRACE(fmt("\n===\nin LHS: %s\nin RHS: %s\n===\n", lhs.to_string().c_str(), rhs.to_string().c_str()));
auto actual = reference_concat(lhs, rhs, "y");
auto expect = perform_simpletensor_concat(lhs, rhs, "y");
EXPECT_EQ(actual, expect);
}
}
TEST(GenericConcatTest, generic_concat_works_for_simple_values) {
ASSERT_TRUE((concat_layouts.size() % 2) == 0);
for (size_t i = 0; i < concat_layouts.size(); i += 2) {
const TensorSpec lhs = spec(concat_layouts[i], N());
const TensorSpec rhs = spec(concat_layouts[i + 1], Div16(N()));
SCOPED_TRACE(fmt("\n===\nin LHS: %s\nin RHS: %s\n===\n", lhs.to_string().c_str(), rhs.to_string().c_str()));
auto actual = perform_generic_concat(lhs, rhs, "y");
auto expect = reference_concat(lhs, rhs, "y");
EXPECT_EQ(actual, expect);
}
}
TEST(GenericConcatTest, dense_concat_plan_can_be_created) {
auto lhs = ValueType::from_spec("tensor(a[2],b[3],c[5],d{},f[2],g[3])");
auto rhs = ValueType::from_spec("tensor(a[2],b[3],c[7],e{},h[3],i[4])");
auto res_type = ValueType::concat(lhs, rhs, "c");
auto plan = DenseConcatPlan(lhs, rhs, "c", res_type);
EXPECT_EQ(plan.right_offset, 5*2*3*3*4);
EXPECT_EQ(plan.output_size, 2*3*12*2*3*3*4);
EXPECT_EQ(plan.left.input_size, 2*3*5*2*3);
std::vector<size_t> expect_left_loop = { 6, 5, 6, 12 };
std::vector<size_t> expect_left_in_s = { 30, 6, 1, 0 };
std::vector<size_t> expect_left_out_s = { 864, 72, 12, 1 };
EXPECT_EQ(plan.left.in_loop_cnt, expect_left_loop);
EXPECT_EQ(plan.left.in_stride, expect_left_in_s);
EXPECT_EQ(plan.left.out_stride, expect_left_out_s);
EXPECT_EQ(plan.right.input_size, 2*3*7*3*4);
std::vector<size_t> expect_right_loop = { 6, 7, 6, 12 };
std::vector<size_t> expect_right_in_s = { 84, 12, 0, 1 };
std::vector<size_t> expect_right_out_s = { 864, 72, 12, 1 };
EXPECT_EQ(plan.right.in_loop_cnt, expect_right_loop);
EXPECT_EQ(plan.right.in_stride, expect_right_in_s);
EXPECT_EQ(plan.right.out_stride, expect_right_out_s);
}
GTEST_MAIN_RUN_ALL_TESTS()
|
