burn-onnx 0.21.0-pre.3

Library for importing ONNX models into the Burn framework
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

use super::prelude::*;

impl NodeCodegen for onnx_ir::lp_pool2d::LpPool2dNode {
    fn inputs(&self) -> &[Argument] {
        &self.inputs
    }

    fn outputs(&self) -> &[Argument] {
        &self.outputs
    }

    fn field(&self) -> Option<Field> {
        let name = Ident::new(&self.name, Span::call_site());
        let kernel_size = self.config.kernel_size.to_tokens();
        let strides = self.config.strides.to_tokens();
        let ceil_mode = self.config.ceil_mode;

        let shape = self.inputs[0].ty.static_shape_known();
        let input_spatial = shape.as_deref().map(|s| &s[2..]);
        let padding = crate::burn::codegen::resolve_auto_pad_2d(
            &self.config.auto_pad,
            &self.config.padding,
            input_spatial,
            &self.config.kernel_size,
            &self.config.strides,
            &self.config.dilation,
        )
        .to_tokens();

        Some(Field::new(
            self.name.clone(),
            quote! {
                AvgPool2d
            },
            quote! {
                let #name = AvgPool2dConfig::new(#kernel_size)
                    .with_strides(#strides)
                    .with_padding(#padding)
                    .with_count_include_pad(true)
                    .with_ceil_mode(#ceil_mode)
                    .init();
            },
        ))
    }

    fn forward(&self, scope: &mut ScopeAtPosition<'_>) -> TokenStream {
        let input = scope.arg(self.inputs.first().unwrap());
        let output = arg_to_ident(self.outputs.first().unwrap());
        let field = Ident::new(&self.name, Span::call_site());

        let p = self.config.p as f32;
        let p_inv = 1.0f32 / p;
        let kernel_size = (self.config.kernel_size[0] * self.config.kernel_size[1]) as f32;

        quote! {
            let #output = self
                .#field
                .forward(#input.abs().powf_scalar(#p))
                .mul_scalar(#kernel_size)
                .powf_scalar(#p_inv);
        }
    }

    fn register_imports(&self, imports: &mut BurnImports) {
        imports.register("burn::nn::pool::AvgPool2d");
        imports.register("burn::nn::pool::AvgPool2dConfig");
        imports.register("burn::nn::PaddingConfig2d");
    }
}

#[cfg(test)]
mod tests {
    use super::super::test_helpers::*;
    use burn::tensor::DType;
    use insta::assert_snapshot;
    use onnx_ir::lp_pool2d::{LpPool2dConfig, LpPool2dNode, LpPool2dNodeBuilder};
    use onnx_ir::padding::{AutoPad, PaddingConfig2d};

    fn create_lp_pool2d_node(name: &str, p: i64) -> LpPool2dNode {
        let config = LpPool2dConfig::new(
            [2, 3],
            [1, 2],
            PaddingConfig2d::Explicit(0, 1, 1, 0),
            [1, 1],
            false,
            AutoPad::NotSet,
            p,
        );

        LpPool2dNodeBuilder::new(name)
            .input_tensor("input", 4, DType::F32)
            .output_tensor("output", 4, DType::F32)
            .config(config)
            .build()
    }

    #[test]
    fn test_lp_pool2d_forward() {
        let node = create_lp_pool2d_node("pool1", 2);
        let code = codegen_forward_default(&node);
        assert_snapshot!(code, @r"
        pub fn forward(&self, input: Tensor<B, 4>) -> Tensor<B, 4> {
            let output = self
                .pool1
                .forward(input.abs().powf_scalar(2f32))
                .mul_scalar(6f32)
                .powf_scalar(0.5f32);
            output
        }
        ");
    }

    #[test]
    fn test_lp_pool2d_forward_with_clone() {
        let node = create_lp_pool2d_node("pool1", 2);
        let code = codegen_forward_with_clone(&node);
        assert_snapshot!(code, @r"
        pub fn forward(&self, input: Tensor<B, 4>) -> Tensor<B, 4> {
            let output = self
                .pool1
                .forward(input.clone().abs().powf_scalar(2f32))
                .mul_scalar(6f32)
                .powf_scalar(0.5f32);
            output
        }
        ");
    }

    #[test]
    fn test_lp_pool2d_field_init() {
        let node = create_lp_pool2d_node("pool1", 2);
        let code = codegen_field_init(&node);
        assert_snapshot!(code, @r#"
        let pool1 = AvgPool2dConfig::new([2, 3])
            .with_strides([1, 2])
            .with_padding(PaddingConfig2d::Explicit(0, 1, 1, 0))
            .with_count_include_pad(true)
            .with_ceil_mode(false)
            .init();
        "#);
    }
}