rotta_rs 0.0.5

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

use crate::rotta_rs_module::{ arrayy::ArrSlice, arrayy::Arrayy, NodeType };

#[derive(Debug, Clone)]
pub enum BackwardLabel {
    // operation
    Dot(NodeType, NodeType),
    Matmul(NodeType, NodeType),
    Add(NodeType, NodeType),
    Diveded(NodeType, NodeType),
    Mul(NodeType, NodeType),
    Sub(NodeType, NodeType),

    // mutation
    Index(NodeType, Vec<i32>),
    Broadcasting(NodeType, Arrayy),
    SumAxis(NodeType, Vec<i32>, bool), // keep_dim
    Sum(NodeType),
    Permute(NodeType, Vec<usize>),
    Slice(NodeType, Vec<ArrSlice>),
    ToShape(NodeType, Vec<usize>),
    Concat(Vec<NodeType>, usize),

    // method
    Exp(NodeType, Arrayy),
    Powi(NodeType, i32),
    Powf(NodeType, f64),
    Ln(NodeType),
    Abs(NodeType),
    Sign(NodeType),

    // activation
    Relu(NodeType),

    // loss
    SSResidual(NodeType, NodeType), // prediction, actual
    CEL(NodeType, NodeType), // prediction, actual
}