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

use super::prelude::*;
use burn_store::TensorSnapshot;

impl NodeCodegen for onnx_ir::node::group_norm::GroupNormalizationNode {
    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 num_groups = self.config.num_groups.to_tokens();
        let scale_shape = self.inputs[1]
            .ty
            .static_shape_known()
            .expect("GroupNorm: scale tensor shape must be known at codegen time");
        let num_features = scale_shape[0].to_tokens();
        let epsilon = self.config.epsilon;

        Some(Field::new(
            self.name.clone(),
            quote! {
                GroupNorm<B>
            },
            quote! {
                let #name = GroupNormConfig::new(#num_groups, #num_features)
                    .with_epsilon(#epsilon)
                    .init(device);
            },
        ))
    }

    fn collect_snapshots(&self, field_name: &str) -> Vec<TensorSnapshot> {
        use crate::burn::node_traits::create_lazy_snapshot;

        let mut snapshots = vec![];

        // Gamma (scale) tensor at input index 1
        if let Some(gamma_input) = self.inputs.get(1) {
            let gamma_path = format!("{}.gamma", field_name);
            if let Some(snapshot) = create_lazy_snapshot(gamma_input, &gamma_path, "GroupNorm") {
                snapshots.push(snapshot);
            }
        }

        // Beta (bias) tensor at input index 2
        if let Some(beta_input) = self.inputs.get(2) {
            let beta_path = format!("{}.beta", field_name);
            if let Some(snapshot) = create_lazy_snapshot(beta_input, &beta_path, "GroupNorm") {
                snapshots.push(snapshot);
            }
        }

        snapshots
    }

    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());

        if self.config.full_precision {
            quote! {
                let #output = {
                    let dtype = #input.dtype();
                    self.#field.forward(#input.cast(burn::tensor::DType::F32)).cast(dtype)
                };
            }
        } else {
            quote! {
                let #output = self.#field.forward(#input);
            }
        }
    }

    fn register_imports(&self, imports: &mut BurnImports) {
        imports.register("burn::nn::GroupNorm");
        imports.register("burn::nn::GroupNormConfig");
    }
}

#[cfg(test)]
mod tests {
    use super::super::test_helpers::*;
    use burn::tensor::DType;
    use insta::assert_snapshot;
    use onnx_ir::node::group_norm::{
        GroupNormConfig, GroupNormalizationNode, GroupNormalizationNodeBuilder,
    };

    fn create_group_norm_node(name: &str) -> GroupNormalizationNode {
        let config = GroupNormConfig::new(8, 1e-5, true);

        GroupNormalizationNodeBuilder::new(name)
            .input_tensor("input", 4, DType::F32)
            .input_static_tensor_shape("scale", vec![64], DType::F32)
            .input_static_tensor_shape("bias", vec![64], DType::F32)
            .output_tensor("output", 4, DType::F32)
            .config(config)
            .build()
    }

    #[test]
    fn test_group_norm_forward() {
        let node = create_group_norm_node("group_norm1");
        let code = codegen_forward_default(&node);
        assert_snapshot!(code, @r"
        pub fn forward(&self, input: Tensor<B, 4>) -> Tensor<B, 4> {
            let output = {
                let dtype = input.dtype();
                self.group_norm1.forward(input.cast(burn::tensor::DType::F32)).cast(dtype)
            };
            output
        }
        ");
    }

    #[test]
    fn test_group_norm_forward_with_clone() {
        let node = create_group_norm_node("group_norm1");
        let code = codegen_forward_with_clone(&node);
        assert_snapshot!(code, @r"
        pub fn forward(&self, input: Tensor<B, 4>) -> Tensor<B, 4> {
            let output = {
                let dtype = input.clone().dtype();
                self.group_norm1
                    .forward(input.clone().cast(burn::tensor::DType::F32))
                    .cast(dtype)
            };
            output
        }
        ");
    }
}