meuron 0.4.0

Meuron is a modular neural network library written in rust for training simple neural networks.
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
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

use bytemuck::{Pod, bytes_of};
use ndarray::Dimension;
use std::sync::Arc;
use wgpu::util::DeviceExt;

use super::context::GpuContext;
use super::params::{BinopParams, ScalarParams, UnaryParams};
use super::tensor::GpuTensor;

pub fn uniform_buf<T: Pod>(ctx: &Arc<GpuContext>, data: &T) -> wgpu::Buffer {
    ctx.device
        .create_buffer_init(&wgpu::util::BufferInitDescriptor {
            label: None,
            contents: bytes_of(data),
            usage: wgpu::BufferUsages::UNIFORM | wgpu::BufferUsages::COPY_DST,
        })
}

pub fn storage_ro_buf<T: Pod>(ctx: &Arc<GpuContext>, data: &T) -> wgpu::Buffer {
    ctx.device
        .create_buffer_init(&wgpu::util::BufferInitDescriptor {
            label: None,
            contents: bytes_of(data),
            usage: wgpu::BufferUsages::STORAGE,
        })
}

fn make_bind_group(
    ctx: &Arc<GpuContext>,
    pipeline: &wgpu::ComputePipeline,
    entries: &[wgpu::BindGroupEntry],
) -> wgpu::BindGroup {
    ctx.device.create_bind_group(&wgpu::BindGroupDescriptor {
        label: None,
        layout: &pipeline.get_bind_group_layout(0),
        entries,
    })
}

pub fn dispatch_1d(
    ctx: &Arc<GpuContext>,
    pipeline: &wgpu::ComputePipeline,
    entries: &[wgpu::BindGroupEntry],
    count: u32,
) {
    let bg = make_bind_group(ctx, pipeline, entries);
    ctx.with_encoder(|enc| {
        let mut pass = enc.begin_compute_pass(&Default::default());
        pass.set_pipeline(pipeline);
        pass.set_bind_group(0, &bg, &[]);
        pass.dispatch_workgroups(count.div_ceil(256), 1, 1);
    });
}

pub fn dispatch_3d(
    ctx: &Arc<GpuContext>,
    pipeline: &wgpu::ComputePipeline,
    entries: &[wgpu::BindGroupEntry],
    x: u32,
    y: u32,
    z: u32,
) {
    let bg = make_bind_group(ctx, pipeline, entries);
    ctx.with_encoder(|enc| {
        let mut pass = enc.begin_compute_pass(&Default::default());
        pass.set_pipeline(pipeline);
        pass.set_bind_group(0, &bg, &[]);
        pass.dispatch_workgroups(x.div_ceil(8), y.div_ceil(8), z);
    });
}

pub fn binop<D: Dimension>(a: &GpuTensor<D>, b: &GpuTensor<D>, op: u32) -> GpuTensor<D> {
    let ctx = a.ctx.clone();
    let size = a.size as u32;
    let out_buf = a.alloc_like();
    let ub = uniform_buf(
        &ctx,
        &BinopParams {
            size,
            op,
            pad0: 0,
            pad1: 0,
        },
    );
    dispatch_1d(
        &ctx,
        &ctx.pipelines.binop,
        &[
            wgpu::BindGroupEntry {
                binding: 0,
                resource: a.buffer.as_entire_binding(),
            },
            wgpu::BindGroupEntry {
                binding: 1,
                resource: b.buffer.as_entire_binding(),
            },
            wgpu::BindGroupEntry {
                binding: 2,
                resource: out_buf.as_entire_binding(),
            },
            wgpu::BindGroupEntry {
                binding: 3,
                resource: ub.as_entire_binding(),
            },
        ],
        size,
    );
    GpuTensor {
        buffer: Arc::new(out_buf),
        shape: a.shape.clone(),
        size: a.size,
        ctx,
    }
}

pub fn scalar_op<D: Dimension>(tensor: &GpuTensor<D>, op: u32, scalar: f32) -> GpuTensor<D> {
    let ctx = tensor.ctx.clone();
    let size = tensor.size as u32;
    let out_buf = tensor.alloc_like();
    let ub = uniform_buf(
        &ctx,
        &ScalarParams {
            size,
            op,
            scalar,
            pad: 0,
        },
    );
    dispatch_1d(
        &ctx,
        &ctx.pipelines.scalar,
        &[
            wgpu::BindGroupEntry {
                binding: 0,
                resource: tensor.buffer.as_entire_binding(),
            },
            wgpu::BindGroupEntry {
                binding: 1,
                resource: out_buf.as_entire_binding(),
            },
            wgpu::BindGroupEntry {
                binding: 2,
                resource: ub.as_entire_binding(),
            },
        ],
        size,
    );
    GpuTensor {
        buffer: Arc::new(out_buf),
        shape: tensor.shape.clone(),
        size: tensor.size,
        ctx,
    }
}

pub fn unary_op<D: Dimension>(tensor: &GpuTensor<D>, op: u32) -> GpuTensor<D> {
    let ctx = tensor.ctx.clone();
    let size = tensor.size as u32;
    let out_buf = tensor.alloc_like();
    let ub = uniform_buf(
        &ctx,
        &UnaryParams {
            size,
            op,
            pad0: 0,
            pad1: 0,
        },
    );
    dispatch_1d(
        &ctx,
        &ctx.pipelines.unary,
        &[
            wgpu::BindGroupEntry {
                binding: 0,
                resource: tensor.buffer.as_entire_binding(),
            },
            wgpu::BindGroupEntry {
                binding: 1,
                resource: out_buf.as_entire_binding(),
            },
            wgpu::BindGroupEntry {
                binding: 2,
                resource: ub.as_entire_binding(),
            },
        ],
        size,
    );
    GpuTensor {
        buffer: Arc::new(out_buf),
        shape: tensor.shape.clone(),
        size: tensor.size,
        ctx,
    }
}