onnx-ir 0.21.0

ONNX-IR is a pure Rust library for parsing ONNX models into an intermediate representation that can be used to generate code for various ML/DL frameworks
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

#!/usr/bin/env -S uv run
# /// script
# dependencies = [
#   "onnx>=1.15.0",
#   "numpy>=1.24.0",
# ]
# ///

"""Generate ONNX model with Gemm that should convert to Linear."""

import onnx
from onnx import helper, TensorProto
import numpy as np


def create_model():
    input_tensor = helper.make_tensor_value_info("input", TensorProto.FLOAT, [2, 3])
    output_tensor = helper.make_tensor_value_info("output", TensorProto.FLOAT, [2, 4])

    # Weight matrix for Gemm
    weight = helper.make_tensor(
        name="weight",
        data_type=TensorProto.FLOAT,
        dims=[3, 4],
        vals=np.random.randn(3, 4).astype(np.float32).tobytes(),
        raw=True,
    )

    # Bias
    bias = helper.make_tensor(
        name="bias",
        data_type=TensorProto.FLOAT,
        dims=[4],
        vals=np.random.randn(4).astype(np.float32).tobytes(),
        raw=True,
    )

    # Gemm: Y = alpha * A * B + beta * C
    # With alpha=1.0, beta=1.0, transA=0, transB=0 → can convert to Linear
    nodes = [
        helper.make_node(
            "Gemm",
            ["input", "weight", "bias"],
            ["output"],
            name="gemm",
            alpha=1.0,
            beta=1.0,
            transA=0,
            transB=0,
        )
    ]

    graph = helper.make_graph(
        nodes,
        "gemm_linear",
        [input_tensor],
        [output_tensor],
        initializer=[weight, bias],
    )
    model = helper.make_model(
        graph, producer_name="onnx-ir-test", opset_imports=[helper.make_opsetid("", 16)]
    )
    onnx.checker.check_model(model)
    return model


if __name__ == "__main__":
    model = create_model()
    onnx.save(model, "../fixtures/gemm_linear.onnx")
    print("Model saved to ../fixtures/gemm_linear.onnx")
    print("  Gemm that can be converted to Linear in Phase 2")