rotta_rs 0.1.0

a Deep Learning library with rust language
Documentation 
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

use crate::rotta_rs_module::{ arrayy::permute_arr, BackwardLabel, Tensor };

pub fn transpose(x: &Tensor, d: (i32, i32)) -> Tensor {
    let arr = x.value.read().unwrap();
    let shape = arr.shape.clone();
    let dimension_t = [d.0, d.1]
        .iter()
        .map(|d| {
            if d >= &(shape.len() as i32) {
                panic!(
                    "range out of array, array is {} dimension but will dimension index is {}",
                    arr.shape.len(),
                    d
                );
            }

            if d < &0 {
                let d = (shape.len() as i32) + d;
                if d < 0 {
                    panic!(
                        "range out of array, array is {} dimension but will dimension index is {}",
                        arr.shape.len(),
                        d
                    );
                }
                d
            } else {
                *d
            }
        })
        .collect::<Vec<i32>>();

    let mut order = shape
        .iter()
        .enumerate()
        .map(|(idx, _)| { idx })
        .collect::<Vec<usize>>();

    // change
    order[dimension_t[0] as usize] = dimension_t[1] as usize;
    order[dimension_t[1] as usize] = dimension_t[0] as usize;

    let arr = permute_arr(&order, &arr);
    let mut tensor = Tensor::from_arrayy(arr);

    tensor.update_parent(vec![x.shared_tensor()]);
    tensor.update_label(Some(BackwardLabel::Permute(x.shared_tensor(), order)));

    tensor
}