tenflowers-core 0.1.1

Core tensor operations and execution engine for TenfloweRS
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
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

// Unary operation compute shaders for f64 type

@group(0) @binding(0) var<storage, read> input: array<f64>;
@group(0) @binding(1) var<storage, read_write> output: array<f64>;

@compute @workgroup_size(64)
fn log_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = log(input[index]);
}

@compute @workgroup_size(64)
fn neg_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = -input[index];
}

@compute @workgroup_size(64)
fn sqrt_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = sqrt(input[index]);
}

@compute @workgroup_size(64)
fn abs_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = abs(input[index]);
}

@compute @workgroup_size(64)
fn exp_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = exp(input[index]);
}

@compute @workgroup_size(64)
fn sin_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = sin(input[index]);
}

@compute @workgroup_size(64)
fn cos_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = cos(input[index]);
}

@compute @workgroup_size(64)
fn tan_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = tan(input[index]);
}

@compute @workgroup_size(64)
fn recip_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = 1.0 / input[index];
}

@compute @workgroup_size(64)
fn floor_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = floor(input[index]);
}

@compute @workgroup_size(64)
fn ceil_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = ceil(input[index]);
}

@compute @workgroup_size(64)
fn round_f64(@builtin(global_invocation_id) global_id: vec3<u32>) {
    let index = global_id.x;
    if (index >= arrayLength(&output)) {
        return;
    }
    output[index] = round(input[index]);
}