test_unfold.py

# Copyright (C) 2018-2025 Intel Corporation
# SPDX-License-Identifier: Apache-2.0

import numpy as np
import pytest
import torch

from pytorch_layer_test_class import PytorchLayerTest


@pytest.mark.parametrize('dimension', (0, 1, 2))
@pytest.mark.parametrize('size', (1, 2))
@pytest.mark.parametrize('step', (1, 2, 3, 4))
@pytest.mark.parametrize('input_shape',
[
    [2, 2, 5], [3, 3, 3, 3], [2, 3, 4, 5]
])
class TestUnfold(PytorchLayerTest):

    def _prepare_input(self):
        return (self.input_tensor, )

    def create_model(self, dimension, size, step):
        class aten_unfold(torch.nn.Module):

            def __init__(self, dimension, size, step) -> None:
                super().__init__()
                self.dimension = dimension
                self.size = size
                self.step = step

            def forward(self, input_tensor):
                return input_tensor.unfold(dimension=self.dimension, size=self.size, step=self.step)

        ref_net = None

        return aten_unfold(dimension, size, step), ref_net, "aten::unfold"

    @pytest.mark.nightly
    @pytest.mark.precommit
    @pytest.mark.precommit_torch_export
    @pytest.mark.precommit_fx_backend
    def test_unfold(self, ie_device, precision, ir_version, dimension, size, step, input_shape):
        self.input_tensor = np.random.randn(*input_shape).astype(np.float32)
        dyn_shape = True
        if ie_device == "GPU" and size == 1 and step == 1:
            dyn_shape = False
        self._test(*self.create_model(dimension, size, step),
                   ie_device, precision, ir_version, dynamic_shapes=dyn_shape)