first commit

tests/models/lw_detr/test_modeling_lw_detr.py

@@ -0,0 +1,815 @@
+# Copyright 2026 The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import unittest
+from functools import cached_property
+
+from transformers import (
+    DeformableDetrImageProcessorPil,
+    LwDetrConfig,
+    LwDetrViTConfig,
+    is_torch_available,
+    is_vision_available,
+)
+from transformers.testing_utils import (
+    Expectations,
+    require_torch,
+    require_vision,
+    slow,
+    torch_device,
+)
+
+from ...test_backbone_common import BackboneTesterMixin
+from ...test_configuration_common import ConfigTester
+from ...test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor
+from ...test_pipeline_mixin import PipelineTesterMixin
+
+
+if is_torch_available():
+    import torch
+    from torch import nn
+
+    from transformers import LwDetrForObjectDetection, LwDetrModel, LwDetrViTBackbone
+
+
+if is_vision_available():
+    from PIL import Image
+
+CHECKPOINT = {
+    "tiny": "stevenbucaille/lwdetr_tiny_30e_objects365",
+    "xlarge": "stevenbucaille/lwdetr_xlarge_30e_objects365",
+}
+
+
+class LwDetrVitModelTester:
+    def __init__(
+        self,
+        parent,
+        batch_size=3,
+        num_labels=3,
+        num_channels=3,
+        use_labels=True,
+        is_training=True,
+        image_size=256,
+        hidden_size=16,
+        num_hidden_layers=2,
+        num_attention_heads=2,
+        window_block_indices=[1],
+        out_indices=[0],
+        num_windows=16,
+        dropout_prob=0.0,
+        attn_implementation="eager",
+    ):
+        self.parent = parent
+        self.batch_size = batch_size
+        self.num_labels = num_labels
+        self.num_channels = num_channels
+        self.use_labels = use_labels
+        self.image_size = image_size
+
+        self.hidden_size = hidden_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.window_block_indices = window_block_indices
+        self.out_indices = out_indices
+        self.num_windows = num_windows
+        self.dropout_prob = dropout_prob
+        self.attn_implementation = attn_implementation
+
+        self.is_training = is_training
+
+    def prepare_config_and_inputs(self):
+        pixel_values = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size])
+
+        labels = None
+        if self.use_labels:
+            labels = ids_tensor([self.batch_size], self.num_labels)
+
+        config = self.get_config()
+
+        return config, pixel_values, labels
+
+    def get_config(self):
+        return LwDetrViTConfig(
+            hidden_size=self.hidden_size,
+            num_hidden_layers=self.num_hidden_layers,
+            num_attention_heads=self.num_attention_heads,
+            window_block_indices=self.window_block_indices,
+            out_indices=self.out_indices,
+            num_windows=self.num_windows,
+            hidden_dropout_prob=self.dropout_prob,
+            attention_probs_dropout_prob=self.dropout_prob,
+            attn_implementation=self.attn_implementation,
+        )
+
+    def prepare_config_and_inputs_for_common(self):
+        config_and_inputs = self.prepare_config_and_inputs()
+        config, pixel_values, labels = config_and_inputs
+        inputs_dict = {"pixel_values": pixel_values}
+        return config, inputs_dict
+
+    def create_and_check_backbone(self, config, pixel_values, labels):
+        model = LwDetrViTBackbone(config=config)
+        model.to(torch_device)
+        model.eval()
+        result = model(pixel_values)
+
+        # verify hidden states
+        self.parent.assertEqual(len(result.feature_maps), len(config.out_features))
+        self.parent.assertListEqual(
+            list(result.feature_maps[0].shape),
+            [
+                self.batch_size,
+                self.hidden_size,
+                self.get_config().num_windows_side ** 2,
+                self.get_config().num_windows_side ** 2,
+            ],
+        )
+
+        # verify channels
+        self.parent.assertEqual(len(model.channels), len(config.out_features))
+        self.parent.assertListEqual(model.channels, [config.hidden_size])
+
+        # verify backbone works with out_features=None
+        config.out_features = None
+        model = LwDetrViTBackbone(config=config)
+        model.to(torch_device)
+        model.eval()
+        result = model(pixel_values)
+
+        # verify feature maps
+        self.parent.assertEqual(len(result.feature_maps), 1)
+        self.parent.assertListEqual(
+            list(result.feature_maps[0].shape),
+            [self.batch_size, config.hidden_size, config.patch_size, config.patch_size],
+        )
+
+        # verify channels
+        self.parent.assertEqual(len(model.channels), 1)
+        self.parent.assertListEqual(model.channels, [config.hidden_size])
+
+
+@require_torch
+class LwDetrViTBackboneTest(ModelTesterMixin, BackboneTesterMixin, unittest.TestCase):
+    all_model_classes = (LwDetrViTBackbone,) if is_torch_available() else ()
+    config_class = LwDetrViTConfig
+    test_resize_embeddings = False
+    test_torch_exportable = True
+    model_split_percents = [0.5, 0.87, 0.9]
+
+    def setUp(self):
+        self.model_tester = LwDetrVitModelTester(self)
+
+    def test_backbone(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_backbone(*config_and_inputs)
+
+    def test_model_get_set_embeddings(self):
+        config, _ = self.model_tester.prepare_config_and_inputs_for_common()
+
+        for model_class in self.all_model_classes:
+            model = model_class(config)
+            self.assertIsInstance(model.get_input_embeddings(), (nn.Module))
+            x = model.get_output_embeddings()
+            self.assertTrue(x is None or isinstance(x, nn.Linear))
+
+    def test_attention_outputs(self):
+        def check_attention_output(inputs_dict, config, model_class):
+            config._attn_implementation = "eager"
+            model = model_class(config)
+            model.to(torch_device)
+            model.eval()
+
+            with torch.no_grad():
+                outputs = model(**self._prepare_for_class(inputs_dict, model_class))
+
+            attentions = outputs.attentions
+
+            windowed_attentions = [attentions[i] for i in self.model_tester.window_block_indices]
+            unwindowed_attentions = [attentions[i] for i in self.model_tester.out_indices]
+
+            expected_windowed_attention_shape = [
+                self.model_tester.batch_size * self.model_tester.num_windows,
+                self.model_tester.num_attention_heads,
+                self.model_tester.get_config().num_windows_side ** 2,
+                self.model_tester.get_config().num_windows_side ** 2,
+            ]
+
+            expected_unwindowed_attention_shape = [
+                self.model_tester.batch_size,
+                self.model_tester.num_attention_heads,
+                self.model_tester.image_size,
+                self.model_tester.image_size,
+            ]
+
+            for i, attention in enumerate(windowed_attentions):
+                self.assertListEqual(
+                    list(attention.shape),
+                    expected_windowed_attention_shape,
+                )
+
+            for i, attention in enumerate(unwindowed_attentions):
+                self.assertListEqual(
+                    list(attention.shape),
+                    expected_unwindowed_attention_shape,
+                )
+
+        config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
+
+        for model_class in self.all_model_classes:
+            inputs_dict["output_attentions"] = True
+            check_attention_output(inputs_dict, config, model_class)
+
+            # check that output_hidden_states also work using config
+            del inputs_dict["output_attentions"]
+            config.output_attentions = True
+
+            check_attention_output(inputs_dict, config, model_class)
+
+    def test_hidden_states_output(self):
+        def check_hidden_states_output(inputs_dict, config, model_class):
+            model = model_class(config)
+            model.to(torch_device)
+            model.eval()
+
+            with torch.no_grad():
+                outputs = model(**self._prepare_for_class(inputs_dict, model_class))
+
+            hidden_states = outputs.hidden_states
+
+            expected_num_stages = self.model_tester.num_hidden_layers
+            self.assertEqual(len(hidden_states), expected_num_stages + 1)
+
+            # VitDet's feature maps are of shape (batch_size, num_channels, height, width)
+            self.assertListEqual(
+                list(hidden_states[0].shape[-2:]),
+                [
+                    self.model_tester.hidden_size,
+                    self.model_tester.hidden_size,
+                ],
+            )
+
+        config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
+
+        for model_class in self.all_model_classes:
+            inputs_dict["output_hidden_states"] = True
+            check_hidden_states_output(inputs_dict, config, model_class)
+
+            # check that output_hidden_states also work using config
+            del inputs_dict["output_hidden_states"]
+            config.output_hidden_states = True
+
+            check_hidden_states_output(inputs_dict, config, model_class)
+
+    # overwrite since LwDetrVitDet only supports retraining gradients of hidden states
+    def test_retain_grad_hidden_states_attentions(self):
+        config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
+        config.output_hidden_states = True
+        config.output_attentions = self.has_attentions
+
+        # no need to test all models as different heads yield the same functionality
+        model_class = self.all_model_classes[0]
+        model = model_class(config)
+        model.to(torch_device)
+
+        inputs = self._prepare_for_class(inputs_dict, model_class)
+
+        outputs = model(**inputs)
+
+        output = outputs.feature_maps[0]
+
+        # Encoder-/Decoder-only models
+        hidden_states = outputs.hidden_states[0]
+        hidden_states.retain_grad()
+
+        output.flatten()[0].backward(retain_graph=True)
+
+        self.assertIsNotNone(hidden_states.grad)
+
+
+def prepare_img():
+    image = Image.open("./tests/fixtures/tests_samples/COCO/000000039769.png")
+    return image
+
+
+class LwDetrModelTester:
+    def __init__(
+        self,
+        parent,
+        batch_size=3,
+        is_training=True,
+        image_size=256,
+        num_labels=5,
+        n_targets=4,
+        use_labels=True,
+        initializer_range=0.02,
+        layer_norm_eps=1e-5,
+        batch_norm_eps=1e-5,
+        # backbone
+        backbone_config=None,
+        # projector
+        projector_scale_factors=[0.5, 2.0],
+        # decoder
+        d_model=32,
+        decoder_ffn_dim=32,
+        decoder_layers=2,
+        decoder_self_attention_heads=2,
+        decoder_cross_attention_heads=4,
+        # model
+        num_queries=10,
+        group_detr=2,
+        dropout=0.0,
+        activation_dropout=0.0,
+        attention_dropout=0.0,
+        attn_implementation="eager",
+    ):
+        self.parent = parent
+        self.batch_size = batch_size
+        self.is_training = is_training
+        self.num_channels = 3
+        self.image_size = image_size
+        self.num_labels = num_labels
+        self.n_targets = n_targets
+        self.use_labels = use_labels
+        self.initializer_range = initializer_range
+        self.layer_norm_eps = layer_norm_eps
+        self.batch_norm_eps = batch_norm_eps
+        self.backbone_config = backbone_config
+        self.projector_scale_factors = projector_scale_factors
+        self.d_model = d_model
+        self.decoder_ffn_dim = decoder_ffn_dim
+        self.decoder_layers = decoder_layers
+        self.decoder_self_attention_heads = decoder_self_attention_heads
+        self.decoder_cross_attention_heads = decoder_cross_attention_heads
+        self.num_queries = num_queries
+        self.group_detr = group_detr
+        self.dropout = dropout
+        self.activation_dropout = activation_dropout
+        self.attention_dropout = attention_dropout
+        self.attn_implementation = attn_implementation
+
+    def prepare_config_and_inputs(self):
+        pixel_values = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size])
+        pixel_mask = torch.ones([self.batch_size, self.image_size, self.image_size], device=torch_device)
+        labels = None
+        if self.use_labels:
+            labels = []
+            for i in range(self.batch_size):
+                target = {}
+                target["class_labels"] = torch.randint(
+                    high=self.num_labels, size=(self.n_targets,), device=torch_device
+                )
+                target["boxes"] = torch.rand(self.n_targets, 4, device=torch_device, dtype=pixel_values.dtype)
+                labels.append(target)
+
+        config = self.get_config()
+        config.num_labels = self.num_labels
+        return config, pixel_values, pixel_mask, labels
+
+    def get_config(self):
+        backbone_config = LwDetrViTConfig(
+            hidden_size=16,
+            num_hidden_layers=4,
+            num_attention_heads=2,
+            window_block_indices=[0, 2],
+            out_indices=[1, 3],
+            num_windows=16,
+            image_size=self.image_size,
+            dropout_prob=self.dropout,
+            attn_implementation=self.attn_implementation,
+        )
+        return LwDetrConfig(
+            backbone_config=backbone_config,
+            d_model=self.d_model,
+            projector_scale_factors=self.projector_scale_factors,
+            decoder_ffn_dim=self.decoder_ffn_dim,
+            decoder_layers=self.decoder_layers,
+            decoder_self_attention_heads=self.decoder_self_attention_heads,
+            decoder_cross_attention_heads=self.decoder_cross_attention_heads,
+            num_queries=self.num_queries,
+            group_detr=self.group_detr,
+            dropout=self.dropout,
+            activation_dropout=self.activation_dropout,
+            attention_dropout=self.attention_dropout,
+            attn_implementation=self.attn_implementation,
+            _attn_implementation=self.attn_implementation,
+        )
+
+    def prepare_config_and_inputs_for_common(self):
+        config, pixel_values, pixel_mask, labels = self.prepare_config_and_inputs()
+        inputs_dict = {"pixel_values": pixel_values, "pixel_mask": pixel_mask}
+        return config, inputs_dict
+
+    def create_and_check_lw_detr_model(self, config, pixel_values, pixel_mask, labels):
+        model = LwDetrModel(config=config)
+        model.to(torch_device)
+        model.eval()
+
+        result = model(pixel_values=pixel_values, pixel_mask=pixel_mask)
+        result = model(pixel_values)
+
+        self.parent.assertEqual(result.last_hidden_state.shape, (self.batch_size, self.num_queries, self.d_model))
+
+    def create_and_check_lw_detr_object_detection_head_model(self, config, pixel_values, pixel_mask, labels):
+        model = LwDetrForObjectDetection(config=config)
+        model.to(torch_device)
+        model.eval()
+
+        result = model(pixel_values=pixel_values, pixel_mask=pixel_mask)
+        result = model(pixel_values)
+
+        self.parent.assertEqual(result.logits.shape, (self.batch_size, self.num_queries, self.num_labels))
+        self.parent.assertEqual(result.pred_boxes.shape, (self.batch_size, self.num_queries, 4))
+
+        result = model(pixel_values=pixel_values, pixel_mask=pixel_mask, labels=labels)
+
+        self.parent.assertEqual(result.loss.shape, ())
+        self.parent.assertEqual(result.logits.shape, (self.batch_size, self.num_queries, self.num_labels))
+        self.parent.assertEqual(result.pred_boxes.shape, (self.batch_size, self.num_queries, 4))
+
+
+@require_torch
+class LwDetrModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase):
+    all_model_classes = (LwDetrModel, LwDetrForObjectDetection) if is_torch_available() else ()
+    pipeline_model_mapping = (
+        {"image-feature-extraction": LwDetrModel, "object-detection": LwDetrForObjectDetection}
+        if is_torch_available()
+        else {}
+    )
+    is_encoder_decoder = False
+    test_missing_keys = False
+    test_torch_exportable = True
+    model_split_percents = [0.5, 0.87, 0.9]
+
+    # special case for head models
+    def _prepare_for_class(self, inputs_dict, model_class, return_labels=False):
+        inputs_dict = super()._prepare_for_class(inputs_dict, model_class, return_labels=return_labels)
+
+        if return_labels:
+            if model_class.__name__ == "LwDetrForObjectDetection":
+                labels = []
+                for i in range(self.model_tester.batch_size):
+                    target = {}
+                    target["class_labels"] = torch.ones(
+                        size=(self.model_tester.n_targets,), device=torch_device, dtype=torch.long
+                    )
+                    target["boxes"] = torch.ones(
+                        self.model_tester.n_targets, 4, device=torch_device, dtype=torch.float
+                    )
+                    labels.append(target)
+                inputs_dict["labels"] = labels
+
+        return inputs_dict
+
+    def setUp(self):
+        self.model_tester = LwDetrModelTester(self)
+        self.config_tester = ConfigTester(
+            self,
+            config_class=LwDetrConfig,
+            has_text_modality=False,
+            common_properties=["d_model", "decoder_self_attention_heads"],
+        )
+
+    def test_config(self):
+        self.config_tester.run_common_tests()
+
+    def test_lw_detr_model(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_lw_detr_model(*config_and_inputs)
+
+    def test_lw_detr_object_detection_head_model(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_lw_detr_object_detection_head_model(*config_and_inputs)
+
+    @unittest.skip(reason="LwDetr does not use inputs_embeds")
+    def test_inputs_embeds(self):
+        pass
+
+    @unittest.skip(reason="LwDetr does not use test_inputs_embeds_matches_input_ids")
+    def test_inputs_embeds_matches_input_ids(self):
+        pass
+
+    @unittest.skip(reason="LwDetr does not support input and output embeddings")
+    def test_model_get_set_embeddings(self):
+        pass
+
+    @unittest.skip(reason="LwDetr does not support input and output embeddings")
+    def test_model_common_attributes(self):
+        pass
+
+    @unittest.skip(reason="LwDetr does not use token embeddings")
+    def test_resize_tokens_embeddings(self):
+        pass
+
+    @unittest.skip(reason="Feed forward chunking is not implemented")
+    def test_feed_forward_chunking(self):
+        pass
+
+    def test_attention_outputs(self):
+        def check_attention_outputs(inputs_dict, config, model_class):
+            model = model_class._from_config(config, attn_implementation="eager")
+            config = model.config
+            model.to(torch_device)
+            model.eval()
+            with torch.no_grad():
+                outputs = model(**self._prepare_for_class(inputs_dict, model_class))
+            attentions = outputs.attentions
+            self.assertEqual(len(attentions), self.model_tester.decoder_layers)
+            expected_attentions_shape = [
+                self.model_tester.batch_size,
+                self.model_tester.decoder_self_attention_heads,
+                self.model_tester.num_queries,
+                self.model_tester.num_queries,
+            ]
+            for i in range(self.model_tester.decoder_layers):
+                self.assertEqual(expected_attentions_shape, list(attentions[i].shape))
+
+            # check cross_attentions outputs
+            expected_attentions_shape = [
+                self.model_tester.batch_size,
+                self.model_tester.num_queries,
+                self.model_tester.decoder_cross_attention_heads,
+                config.num_feature_levels,
+                config.decoder_n_points,
+            ]
+            cross_attentions = outputs.cross_attentions
+            self.assertEqual(len(cross_attentions), self.model_tester.decoder_layers)
+            for i in range(self.model_tester.decoder_layers):
+                self.assertEqual(expected_attentions_shape, list(cross_attentions[i].shape))
+
+            out_len = len(outputs)
+
+            correct_outlen = 8  # 6 + attentions + cross_attentions
+
+            # Object Detection model returns pred_logits, pred_boxes and auxiliary outputs
+            if model_class.__name__ == "LwDetrForObjectDetection":
+                correct_outlen += 2
+                if "labels" in inputs_dict:
+                    correct_outlen += 3  # loss, loss_dict and auxiliary outputs is added to beginning
+
+            self.assertEqual(correct_outlen, out_len)
+
+        config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
+        config.return_dict = True
+        inputs_dict["output_hidden_states"] = False
+
+        for model_class in self.all_model_classes:
+            inputs_dict["output_attentions"] = True
+            check_attention_outputs(inputs_dict, config, model_class)
+
+            # check that output_attentions also work using config
+            del inputs_dict["output_attentions"]
+            config.output_attentions = True
+            check_attention_outputs(inputs_dict, config, model_class)
+
+    def test_hidden_states_output(self):
+        def check_hidden_states_output(inputs_dict, config, model_class):
+            model = model_class(config)
+            model.to(torch_device)
+            model.eval()
+
+            with torch.no_grad():
+                outputs = model(**self._prepare_for_class(inputs_dict, model_class))
+
+            hidden_states = outputs.hidden_states
+
+            expected_num_hidden_states = self.model_tester.decoder_layers + 1
+            self.assertEqual(len(hidden_states), expected_num_hidden_states)
+
+            for i in range(expected_num_hidden_states):
+                self.assertListEqual(
+                    list(hidden_states[i].shape),
+                    [
+                        self.model_tester.batch_size,
+                        self.model_tester.num_queries,
+                        self.model_tester.d_model,
+                    ],
+                )
+
+        config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
+
+        for model_class in self.all_model_classes:
+            inputs_dict["output_attentions"] = False
+            inputs_dict["output_hidden_states"] = True
+            check_hidden_states_output(inputs_dict, config, model_class)
+
+            # check that output_hidden_states also work using config
+            del inputs_dict["output_hidden_states"]
+            config.output_hidden_states = True
+            check_hidden_states_output(inputs_dict, config, model_class)
+
+    def test_retain_grad_hidden_states_attentions(self):
+        config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
+        config.output_hidden_states = True
+        config.output_attentions = True
+
+        model_class = self.all_model_classes[0]
+        model = model_class(config)
+        model.to(torch_device)
+
+        inputs = self._prepare_for_class(inputs_dict, model_class)
+
+        outputs = model(**inputs)
+
+        # we take the first output since last_hidden_state is the first item
+        output = outputs.last_hidden_state
+
+        hidden_states = outputs.hidden_states[0]
+        attentions = outputs.attentions[0]
+        hidden_states.retain_grad()
+        attentions.retain_grad()
+
+        output.flatten()[0].backward(retain_graph=True)
+
+        self.assertIsNotNone(hidden_states.grad)
+        self.assertIsNotNone(attentions.grad)
+
+    def test_forward_auxiliary_loss(self):
+        config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
+        config.auxiliary_loss = True
+
+        # only test for object detection and segmentation model
+        for model_class in self.all_model_classes[1:]:
+            model = model_class(config)
+            model.to(torch_device)
+
+            inputs = self._prepare_for_class(inputs_dict, model_class, return_labels=True)
+
+            outputs = model(**inputs)
+
+            self.assertIsNotNone(outputs.auxiliary_outputs)
+            self.assertEqual(len(outputs.auxiliary_outputs), self.model_tester.decoder_layers - 1)
+
+
+@require_torch
+@require_vision
+class LwDetrModelIntegrationTest(unittest.TestCase):
+    @cached_property
+    def default_image_processor(self):
+        if is_vision_available():
+            return {
+                "tiny": DeformableDetrImageProcessorPil.from_pretrained(CHECKPOINT["tiny"]),
+                "xlarge": DeformableDetrImageProcessorPil.from_pretrained(CHECKPOINT["xlarge"]),
+            }
+
+    @slow
+    def test_inference_object_detection_head_tiny(self):
+        size = "tiny"
+        model = LwDetrForObjectDetection.from_pretrained(CHECKPOINT[size], attn_implementation="eager").to(
+            torch_device
+        )
+
+        image_processor = self.default_image_processor[size]
+        image = prepare_img()
+        encoding = image_processor(images=image, return_tensors="pt").to(torch_device)
+        pixel_values = encoding["pixel_values"].to(torch_device)
+        pixel_mask = encoding["pixel_mask"].to(torch_device)
+        with torch.no_grad():
+            outputs = model(pixel_values, pixel_mask)
+
+        expected_logits_shape = torch.Size((1, model.config.num_queries, model.config.num_labels))
+        self.assertEqual(outputs.logits.shape, expected_logits_shape)
+
+        expectations = Expectations(
+            {
+                ("cuda", (8, 0)): [-7.7648, -4.1330, -2.9003, -4.0559, -2.9635],
+                ("xpu", (3, 0)): [-7.7693, -4.1270, -2.9018, -4.0605, -2.9575],
+            }
+        )
+        expected_logits = torch.tensor(expectations.get_expectation()).to(torch_device)
+        torch.testing.assert_close(outputs.logits.flatten()[:5], expected_logits, rtol=2e-4, atol=2e-4)
+
+        expected_boxes_shape = torch.Size((1, model.config.num_queries, 4))
+        self.assertEqual(outputs.pred_boxes.shape, expected_boxes_shape)
+
+        expectations = Expectations(
+            {
+                ("cuda", (8, 0)): [0.1694, 0.1979, 0.2121, 0.0912, 0.2537],
+                ("xpu", (3, 0)): [0.1694, 0.1979, 0.2121, 0.0912, 0.2537],
+            }
+        )
+        expected_boxes = torch.tensor(expectations.get_expectation()).to(torch_device)
+
+        torch.testing.assert_close(outputs.pred_boxes.flatten()[:5], expected_boxes, rtol=2e-4, atol=2e-4)
+
+        results = image_processor.post_process_object_detection(
+            outputs, threshold=0.0, target_sizes=[image.size[::-1]]
+        )[0]
+
+        expectations = Expectations(
+            {
+                ("cuda", (8, 0)): [0.8684, 0.7492, 0.7146, 0.4362],
+                ("xpu", (3, 0)): [0.8676, 0.7527, 0.7177, 0.4391],
+            }
+        )
+        expected_scores = torch.tensor(expectations.get_expectation()).to(torch_device)
+
+        expected_labels = [140, 133, 140, 133]
+
+        expectations = Expectations(
+            {
+                ("cuda", (8, 0)): [
+                    [4.9333, 56.6130, 319.7758, 474.7774],
+                    [40.5547, 73.0968, 176.2951, 116.8605],
+                    [340.3403, 25.1044, 640.2798, 368.7382],
+                    [334.2971, 77.0087, 371.2877, 189.8089],
+                ],
+                ("xpu", (3, 0)): [
+                    [4.8948, 56.5549, 319.8077, 474.7937],
+                    [40.5620, 73.1059, 176.2996, 116.8567],
+                    [340.3327, 25.1026, 640.3193, 368.6754],
+                    [334.2945, 76.9876, 371.2914, 189.8221],
+                ],
+            }
+        )
+        expected_slice_boxes = torch.tensor(expectations.get_expectation()).to(torch_device)
+        torch.testing.assert_close(results["scores"][:4], expected_scores, atol=1e-3, rtol=2e-4)
+        self.assertSequenceEqual(results["labels"][:4].tolist(), expected_labels)
+        torch.testing.assert_close(results["boxes"][:4], expected_slice_boxes, atol=1e-3, rtol=2e-4)
+
+    @slow
+    def test_inference_object_detection_head_xlarge(self):
+        size = "xlarge"
+        model = LwDetrForObjectDetection.from_pretrained(CHECKPOINT[size], attn_implementation="eager").to(
+            torch_device
+        )
+
+        image_processor = self.default_image_processor[size]
+        image = prepare_img()
+        encoding = image_processor(images=image, return_tensors="pt").to(torch_device)
+        pixel_values = encoding["pixel_values"].to(torch_device)
+        pixel_mask = encoding["pixel_mask"].to(torch_device)
+        with torch.no_grad():
+            outputs = model(pixel_values, pixel_mask)
+
+        expected_logits_shape = torch.Size((1, model.config.num_queries, model.config.num_labels))
+        self.assertEqual(outputs.logits.shape, expected_logits_shape)
+
+        expectations = Expectations(
+            {
+                ("cuda", (8, 0)): [-11.9394, -4.3419, -4.4172, -5.0299, -6.9282],
+                ("xpu", (3, 0)): [-11.9292, -4.3307, -4.4075, -5.0207, -6.9211],
+            }
+        )
+
+        expected_logits = torch.tensor(expectations.get_expectation()).to(torch_device)
+        torch.testing.assert_close(outputs.logits.flatten()[:5], expected_logits, rtol=2e-4, atol=2e-4)
+
+        expected_boxes_shape = torch.Size((1, model.config.num_queries, 4))
+        self.assertEqual(outputs.pred_boxes.shape, expected_boxes_shape)
+
+        expectations = Expectations(
+            {
+                ("cuda", (8, 0)): [0.7689, 0.4107, 0.4617, 0.7244, 0.2526],
+                ("xpu", (3, 0)): [0.7688, 0.4106, 0.4618, 0.7245, 0.2526],
+            }
+        )
+        expected_boxes = torch.tensor(expectations.get_expectation()).to(torch_device)
+
+        torch.testing.assert_close(outputs.pred_boxes.flatten()[:5], expected_boxes, rtol=2e-4, atol=2e-4)
+
+        results = image_processor.post_process_object_detection(
+            outputs, threshold=0.0, target_sizes=[image.size[::-1]]
+        )[0]
+
+        expectations = Expectations(
+            {
+                ("cuda", (8, 0)): [0.9746, 0.9717, 0.9344, 0.8182],
+                ("xpu", (3, 0)): [0.9745, 0.9715, 0.9339, 0.8163],
+            }
+        )
+        expected_scores = torch.tensor(expectations.get_expectation()).to(torch_device)
+
+        expected_labels = [140, 140, 133, 133]
+
+        expectations = Expectations(
+            {
+                ("cuda", (8, 0)): [
+                    [7.4541, 54.2878, 315.8890, 474.8681],
+                    [344.3325, 23.2591, 639.7999, 370.9900],
+                    [40.4797, 73.3092, 175.6086, 116.9654],
+                    [333.9930, 77.1547, 370.4000, 186.1230],
+                ],
+                ("xpu", (3, 0)): [
+                    [7.4487, 54.2931, 315.8945, 474.8726],
+                    [344.2597, 23.2305, 639.8082, 370.9894],
+                    [40.4780, 73.3095, 175.6083, 116.9673],
+                    [333.9890, 77.1453, 370.4069, 186.1300],
+                ],
+            }
+        )
+        expected_slice_boxes = torch.tensor(expectations.get_expectation()).to(torch_device)
+
+        torch.testing.assert_close(results["scores"][:4], expected_scores, atol=1e-3, rtol=2e-4)
+        self.assertSequenceEqual(results["labels"][:4].tolist(), expected_labels)
+        torch.testing.assert_close(results["boxes"][:4], expected_slice_boxes, atol=1e-3, rtol=2e-4)