ella_tensor/ops/
constructors.rs

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
use arrow::array::Array;
use num_traits::{Float, Num, One, Zero};

use crate::{Const, IntoShape, Shape, Tensor, TensorValue};

impl<T, S> Tensor<T, S>
where
    T: TensorValue,
    S: Shape,
{
    pub fn full<I, V>(shape: I, value: V) -> Self
    where
        I: IntoShape<Shape = S>,
        V: Into<T>,
    {
        let shape = shape.into_shape();
        unsafe {
            Tensor::from_trusted_len_iter(std::iter::repeat(value.into()).take(shape.size()), shape)
        }
    }

    pub fn zeros<I>(shape: I) -> Self
    where
        I: IntoShape<Shape = S>,
        T: Zero,
    {
        Self::full(shape, T::zero())
    }

    pub fn ones<I>(shape: I) -> Self
    where
        I: IntoShape<Shape = S>,
        T: One,
    {
        Self::full(shape, T::one())
    }

    pub(crate) unsafe fn from_trusted_len_iter<I>(iter: I, shape: S) -> Self
    where
        I: IntoIterator<Item = T>,
    {
        let values = unsafe { T::from_trusted_len_iter(iter) };
        let strides = shape.default_strides();
        Tensor::new(values, shape, strides)
    }
}

/// 1-D constructors
impl<T> Tensor<T, Const<1>>
where
    T: TensorValue,
{
    pub fn linspace(start: T, end: T, steps: usize) -> Self
    where
        T: Float,
    {
        let step_size = (end - start) / T::from(steps).unwrap();
        let values = (0..steps).map(|i| start + step_size * T::from(i).unwrap());
        let shape = Const([steps]);
        unsafe { Tensor::from_trusted_len_iter(values, shape) }
    }

    pub fn range(start: T, end: T, step: T) -> Self
    where
        T: Num,
    {
        let values = std::iter::successors(Some(start), |x| {
            let value = x.clone() + step.clone();
            if value < end {
                Some(value)
            } else {
                None
            }
        });
        values.collect()
    }
}

/// 2-D constructors
impl<T> Tensor<T, Const<2>>
where
    T: TensorValue,
{
    pub fn eye(size: usize) -> Self
    where
        T: One + Zero,
    {
        let shape = Const([size, size]);
        let iter = shape
            .indices()
            .map(|Const([r, c])| if r == c { T::one() } else { T::zero() });
        unsafe { Tensor::from_trusted_len_iter(iter, shape) }
    }
}

impl<T> From<Vec<T>> for Tensor<T, Const<1>>
where
    T: TensorValue,
{
    fn from(value: Vec<T>) -> Self {
        let shape = Const([value.len()]);
        let strides = shape.default_strides();
        Tensor::new(T::from_vec(value), shape, strides)
    }
}

impl<T, const D: usize> From<Vec<[T; D]>> for Tensor<T, Const<2>>
where
    T: TensorValue,
{
    fn from(value: Vec<[T; D]>) -> Self {
        let shape = Const([value.len(), D]);
        let strides = shape.default_strides();
        let values = unsafe { T::from_trusted_len_iter(value.into_iter().flatten()) };
        Tensor::new(values, shape, strides)
    }
}

impl<T, const D1: usize, const D2: usize> From<Vec<[[T; D2]; D1]>> for Tensor<T, Const<3>>
where
    T: TensorValue,
{
    fn from(value: Vec<[[T; D2]; D1]>) -> Self {
        let shape = Const([value.len(), D1, D2]);
        let strides = shape.default_strides();
        let values = unsafe {
            T::from_trusted_len_iter(
                value.into_iter().flatten().flatten(), // .flat_map(|inner| inner.into_iter().flatten()),
            )
        };
        Tensor::new(values, shape, strides)
    }
}

impl<T> From<T> for Tensor<T, Const<0>>
where
    T: TensorValue,
{
    fn from(value: T) -> Self {
        let values = unsafe { T::from_trusted_len_iter(std::iter::once(value)) };
        let shape = Const([]);
        let strides = Const([]);

        Tensor::new(values, shape, strides)
    }
}

impl<T> FromIterator<T> for Tensor<T, Const<1>>
where
    T: TensorValue,
{
    fn from_iter<I: IntoIterator<Item = T>>(iter: I) -> Self {
        let values = T::from_iter(iter);
        let shape = Const([values.len()]);
        let strides = shape.default_strides();
        Tensor::new(values, shape, strides)
    }
}