-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy pathTensor.hpp
More file actions
210 lines (169 loc) · 5.65 KB
/
Copy pathTensor.hpp
File metadata and controls
210 lines (169 loc) · 5.65 KB
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
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
#pragma once
#include <cstdint>
#include <iostream>
#include <stdexcept>
#include <type_traits>
#include <vector>
#include "layers/Shape.hpp"
namespace it_lab_ai {
enum class Type : uint8_t { kUnknown, kInt, kFloat };
template <typename T>
std::vector<uint8_t>* to_byte(std::vector<T>& v) {
return reinterpret_cast<std::vector<uint8_t>*>(&v);
}
template <typename T>
const std::vector<uint8_t>* to_byte(const std::vector<T>& v) {
return reinterpret_cast<const std::vector<uint8_t>*>(&v);
}
template <typename T>
Type GetTypeEnum() {
if constexpr (std::is_same_v<T, int>) {
return Type::kInt;
} else if constexpr (std::is_same_v<T, float>) {
return Type::kFloat;
} else {
return Type::kUnknown;
}
}
class Tensor {
private:
Shape shape_;
std::vector<uint8_t> values_;
std::vector<float> bias_;
Type type_;
std::vector<uint8_t> SetRightTypeValues() {
if (type_ == Type::kInt) {
return std::vector<uint8_t>(shape_.count() * sizeof(int), 0);
}
if (type_ == Type::kFloat) {
return std::vector<uint8_t>(shape_.count() * sizeof(float), 0);
}
return std::vector<uint8_t>();
}
public:
Tensor() = default;
Tensor(const std::vector<uint8_t>& a, const Shape& sh, Type type)
: shape_(sh), type_(type) {
if (type == Type::kUnknown) {
throw std::invalid_argument("Unknown data type");
}
values_ = SetRightTypeValues();
if (a.size() != values_.size()) {
throw std::invalid_argument("Incorrect vector size given to Tensor");
}
values_ = a;
}
Tensor(const Shape& sh, Type type) : shape_(sh), type_(type) {
if (type == Type::kUnknown) {
throw std::invalid_argument("Unknown data type");
}
values_ = SetRightTypeValues();
}
Tensor(const std::vector<uint8_t>& a, const Shape& sh,
const std::vector<float>& bias)
: shape_(sh), values_(a), bias_(bias), type_(Type::kFloat) {
values_ = SetRightTypeValues();
if (a.size() != values_.size()) {
throw std::invalid_argument("Incorrect vector size given to Tensor");
}
values_ = a;
}
Tensor(const Tensor& t) = default;
Tensor(Tensor&& t) = default;
Tensor& operator=(Tensor&& t) = default;
Tensor& operator=(const Tensor& t) = default;
Shape get_shape() const { return shape_; }
Type get_type() const noexcept { return type_; }
void set_bias(const std::vector<float>& bias) {
if (bias.size() != shape_[shape_.dims() - 1]) {
throw std::invalid_argument(
"Bias size does not match the last dimension of the shape");
}
bias_ = bias;
}
const std::vector<float>& get_bias() const { return bias_; }
const std::vector<uint8_t>& get_values() const { return values_; }
bool empty() const { return values_.empty(); }
auto begin() { return values_.begin(); }
auto end() { return values_.end(); }
auto begin() const { return values_.begin(); }
auto end() const { return values_.end(); }
template <typename T>
typename std::vector<T>::const_iterator begin() const {
return this->as<T>().begin();
}
template <typename T>
typename std::vector<T>::const_iterator end() const {
return this->as<T>().end();
}
template <typename T>
void set(const std::vector<size_t>& coords, const T& elem);
template <typename T>
T get(const std::vector<size_t>& coords) const;
template <typename T>
std::vector<T>* as();
template <typename T>
const std::vector<T>* as() const;
friend std::ostream& operator<<(std::ostream& out, const Tensor& t);
};
template <typename T>
void Tensor::set(const std::vector<size_t>& coords, const T& elem) {
size_t s = shape_.get_index(coords);
std::vector<T>* res_vector = this->as<T>();
if ((*res_vector).size() == 0) {
throw std::invalid_argument("Empty tensor");
}
(*res_vector)[s] = elem;
}
template <typename T>
T Tensor::get(const std::vector<size_t>& coords) const {
size_t s = shape_.get_index(coords);
const std::vector<T>* res_vector = this->as<T>();
if ((*res_vector).size() == 0) {
throw std::invalid_argument("Empty tensor");
}
return (*res_vector)[s];
}
template <typename T>
std::vector<T>* Tensor::as() {
if (GetTypeEnum<T>() != type_) {
throw std::invalid_argument("Template type doesn't fit this Tensor");
}
return reinterpret_cast<std::vector<T>*>(&values_);
}
template <typename T>
const std::vector<T>* Tensor::as() const {
if (GetTypeEnum<T>() != type_) {
throw std::invalid_argument("Template type doesn't fit this Tensor");
}
return reinterpret_cast<const std::vector<T>*>(&values_);
}
template <typename T>
Tensor make_tensor(const std::vector<T>& values) {
Type type = GetTypeEnum<T>();
if (type == Type::kUnknown) {
throw std::invalid_argument("Unsupported tensor type");
}
Shape shape({values.size()});
std::vector<uint8_t> byte_values(
reinterpret_cast<const uint8_t*>(values.data()),
reinterpret_cast<const uint8_t*>(values.data() + values.size()));
return Tensor(byte_values, shape, type);
}
template <typename T>
Tensor make_tensor(const std::vector<T>& values, const Shape& shape) {
std::vector<uint8_t> byte_values(
reinterpret_cast<const uint8_t*>(values.data()),
reinterpret_cast<const uint8_t*>(values.data() + values.size()));
return Tensor(byte_values, shape, GetTypeEnum<T>());
}
template <typename T>
Tensor make_tensor(const std::vector<T>& values, const Shape& shape,
const std::vector<float>& bias) {
std::vector<uint8_t> byte_values(
reinterpret_cast<const uint8_t*>(values.data()),
reinterpret_cast<const uint8_t*>(values.data() + values.size()));
return Tensor(byte_values, shape, bias);
}
std::ostream& operator<<(std::ostream& out, const Tensor& t);
} // namespace it_lab_ai