mmdeploy/tests/test_apis/test_torch2onnx.py

# Copyright (c) OpenMMLab. All rights reserved.
import os.path as osp
import tempfile

import mmcv
import onnx
import pytest
import torch
import torch.nn as nn

from mmdeploy.apis.onnx import export
from mmdeploy.utils.config_utils import (get_backend, get_dynamic_axes,
                                         get_onnx_config)
from mmdeploy.utils.test import get_random_name

onnx_file = tempfile.NamedTemporaryFile(suffix='.onnx').name


@pytest.mark.skip(reason='This a not test class but a utility class.')
class TestModel(nn.Module):

    def __init__(self):
        super().__init__()

    def forward(self, x):
        return x * 0.5


test_model = TestModel().eval().cuda()
test_img = torch.rand([1, 3, 8, 8])
input_name = get_random_name()
output_name = get_random_name()
dynamic_axes_dict = {
    input_name: {
        0: 'batch',
        2: 'height',
        3: 'width'
    },
    output_name: {
        0: 'batch'
    }
}
dynamic_axes_list = [[0, 2, 3], [0]]


def get_deploy_cfg(input_name, output_name, dynamic_axes):
    return mmcv.Config(
        dict(
            onnx_config=dict(
                dynamic_axes=dynamic_axes,
                type='onnx',
                export_params=True,
                keep_initializers_as_inputs=False,
                opset_version=11,
                input_names=[input_name],
                output_names=[output_name],
                input_shape=None),
            codebase_config=dict(type='mmedit', task=''),
            backend_config=dict(type='onnxruntime')))


@pytest.mark.parametrize('input_name', [input_name])
@pytest.mark.parametrize('output_name', [output_name])
@pytest.mark.parametrize('dynamic_axes',
                         [dynamic_axes_dict, dynamic_axes_list])
def test_torch2onnx(input_name, output_name, dynamic_axes):
    deploy_cfg = get_deploy_cfg(input_name, output_name, dynamic_axes)

    output_prefix = osp.splitext(onnx_file)[0]
    context_info = dict(cfg=deploy_cfg)
    backend = get_backend(deploy_cfg).value
    onnx_cfg = get_onnx_config(deploy_cfg)
    opset_version = onnx_cfg.get('opset_version', 11)

    input_names = onnx_cfg['input_names']
    output_names = onnx_cfg['output_names']
    axis_names = input_names + output_names
    dynamic_axes = get_dynamic_axes(deploy_cfg, axis_names)
    verbose = not onnx_cfg.get('strip_doc_string', True) or onnx_cfg.get(
        'verbose', False)
    keep_initializers_as_inputs = onnx_cfg.get('keep_initializers_as_inputs',
                                               True)
    export(
        test_model,
        test_img,
        context_info=context_info,
        output_path_prefix=output_prefix,
        backend=backend,
        input_names=input_names,
        output_names=output_names,
        opset_version=opset_version,
        dynamic_axes=dynamic_axes,
        verbose=verbose,
        keep_initializers_as_inputs=keep_initializers_as_inputs)

    assert osp.exists(onnx_file)

    model = onnx.load(onnx_file)
    assert model is not None
    try:
        onnx.checker.check_model(model)
    except onnx.checker.ValidationError:
        assert False