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
use crate::{backend::Backend, tensor::Shape, Data, ElementConversion};
use std::ops::Range;
pub trait ModuleOps<B: Backend> {
fn embedding(
weights: &B::TensorPrimitive<2>,
indexes: &<B::IntegerBackend as Backend>::TensorPrimitive<2>,
) -> B::TensorPrimitive<3>;
fn embedding_backward(
weights: &B::TensorPrimitive<2>,
output: &B::TensorPrimitive<3>,
indexes: &<B::IntegerBackend as Backend>::TensorPrimitive<2>,
) -> B::TensorPrimitive<2>;
}
pub trait TensorOps<B: Backend> {
fn shape<const D: usize>(tensor: &B::TensorPrimitive<D>) -> &Shape<D>;
fn to_data<const D: usize>(tensor: &B::TensorPrimitive<D>) -> Data<B::Elem, D>;
fn into_data<const D: usize>(tensor: B::TensorPrimitive<D>) -> Data<B::Elem, D>;
fn bool_shape<const D: usize>(tensor: &B::BoolTensorPrimitive<D>) -> &Shape<D>;
fn bool_to_data<const D: usize>(tensor: &B::BoolTensorPrimitive<D>) -> Data<bool, D>;
fn bool_into_data<const D: usize>(tensor: B::BoolTensorPrimitive<D>) -> Data<bool, D>;
fn device<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::Device;
fn to_device<const D: usize>(
tensor: &B::TensorPrimitive<D>,
device: B::Device,
) -> B::TensorPrimitive<D>;
fn arange(
range: Range<usize>,
device: B::Device,
) -> <B::IntegerBackend as Backend>::TensorPrimitive<1> {
let shape = Shape::new([range.end - range.start]);
let value = range
.into_iter()
.map(|i| (i as i64).to_elem())
.collect::<Vec<<B::IntegerBackend as Backend>::Elem>>();
let data = Data::new(value, shape);
<B::IntegerBackend as Backend>::from_data(data, device)
}
fn empty<const D: usize>(shape: Shape<D>, device: B::Device) -> B::TensorPrimitive<D>;
fn repeat<const D: usize>(
tensor: &B::TensorPrimitive<D>,
dim: usize,
times: usize,
) -> B::TensorPrimitive<D> {
let mut shape = *B::shape(tensor);
if shape.dims[dim] != 1 {
panic!("Can only repeat dimension with dim=1");
}
shape.dims[dim] = times;
let mut i = 0;
let indexes_select_all = [0; D].map(|_| {
let start = 0;
let end = shape.dims[i];
i += 1;
start..end
});
let mut tensor_output = B::empty(shape, B::device(tensor));
for i in 0..times {
let mut indexes = indexes_select_all.clone();
indexes[dim] = i..i + 1;
tensor_output = B::index_assign(&tensor_output, indexes, tensor);
}
tensor_output
}
fn add<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::TensorPrimitive<D>;
fn add_scalar<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::Elem,
) -> B::TensorPrimitive<D>;
fn sub<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::TensorPrimitive<D>;
fn sub_scalar<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::Elem,
) -> B::TensorPrimitive<D>;
fn mul<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::TensorPrimitive<D>;
fn mul_scalar<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::Elem,
) -> B::TensorPrimitive<D>;
fn div<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::TensorPrimitive<D>;
fn div_scalar<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::Elem,
) -> B::TensorPrimitive<D>;
fn matmul<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::TensorPrimitive<D>;
fn neg<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::TensorPrimitive<D>;
fn transpose<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::TensorPrimitive<D> {
Self::swap_dims(tensor, D - 2, D - 1)
}
fn swap_dims<const D: usize>(
tensor: &B::TensorPrimitive<D>,
dim1: usize,
dim2: usize,
) -> B::TensorPrimitive<D>;
fn reshape<const D1: usize, const D2: usize>(
tensor: &B::TensorPrimitive<D1>,
shape: Shape<D2>,
) -> B::TensorPrimitive<D2>;
fn index<const D1: usize, const D2: usize>(
tensor: &B::TensorPrimitive<D1>,
indexes: [Range<usize>; D2],
) -> B::TensorPrimitive<D1>;
fn index_assign<const D1: usize, const D2: usize>(
tensor: &B::TensorPrimitive<D1>,
indexes: [Range<usize>; D2],
value: &B::TensorPrimitive<D1>,
) -> B::TensorPrimitive<D1>;
fn mask_fill<const D: usize>(
tensor: &B::TensorPrimitive<D>,
mask: &B::BoolTensorPrimitive<D>,
value: B::Elem,
) -> B::TensorPrimitive<D>;
fn equal<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::BoolTensorPrimitive<D>;
fn equal_scalar<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::Elem,
) -> B::BoolTensorPrimitive<D>;
fn greater<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::BoolTensorPrimitive<D>;
fn greater_scalar<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::Elem,
) -> B::BoolTensorPrimitive<D>;
fn greater_equal<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::BoolTensorPrimitive<D>;
fn greater_equal_scalar<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::Elem,
) -> B::BoolTensorPrimitive<D>;
fn lower<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::BoolTensorPrimitive<D>;
fn lower_scalar<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::Elem,
) -> B::BoolTensorPrimitive<D>;
fn lower_equal<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::TensorPrimitive<D>,
) -> B::BoolTensorPrimitive<D>;
fn lower_equal_scalar<const D: usize>(
lhs: &B::TensorPrimitive<D>,
rhs: &B::Elem,
) -> B::BoolTensorPrimitive<D>;
fn detach<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::TensorPrimitive<D>;
fn mean<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::TensorPrimitive<1>;
fn sum<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::TensorPrimitive<1>;
fn mean_dim<const D: usize>(
tensor: &B::TensorPrimitive<D>,
dim: usize,
) -> B::TensorPrimitive<D>;
fn sum_dim<const D: usize>(tensor: &B::TensorPrimitive<D>, dim: usize)
-> B::TensorPrimitive<D>;
fn to_full_precision<const D: usize>(
tensor: &B::TensorPrimitive<D>,
) -> <B::FullPrecisionBackend as Backend>::TensorPrimitive<D>;
fn from_full_precision<const D: usize>(
tensor: &<B::FullPrecisionBackend as Backend>::TensorPrimitive<D>,
) -> B::TensorPrimitive<D>;
fn argmax<const D: usize>(
tensor: &B::TensorPrimitive<D>,
dim: usize,
) -> <B::IntegerBackend as Backend>::TensorPrimitive<D>;
fn argmin<const D: usize>(
tensor: &B::TensorPrimitive<D>,
dim: usize,
) -> <B::IntegerBackend as Backend>::TensorPrimitive<D>;
fn exp<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::TensorPrimitive<D>;
fn log<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::TensorPrimitive<D>;
fn powf<const D: usize>(tensor: &B::TensorPrimitive<D>, value: f32) -> B::TensorPrimitive<D>;
fn erf<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::TensorPrimitive<D>;
fn cat<const D: usize>(tensors: &[B::TensorPrimitive<D>], dim: usize) -> B::TensorPrimitive<D>;
fn relu<const D: usize>(tensor: &B::TensorPrimitive<D>) -> B::TensorPrimitive<D>;
}
pub trait Zeros {
fn zeros(&self) -> Self;
}
pub trait Ones {
fn ones(&self) -> Self;
}