first commit

tests/models/pp_chart2table/test_image_processing_pp_chart2table.py

@@ -0,0 +1,94 @@
+# 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 transformers.testing_utils import require_torch, require_vision
+
+from ...test_image_processing_common import ImageProcessingTestMixin, prepare_image_inputs
+
+
+class PPChart2TableImageProcessingTester(unittest.TestCase):
+    def __init__(
+        self,
+        parent,
+        batch_size=7,
+        num_channels=3,
+        image_size=18,
+        min_resolution=30,
+        max_resolution=400,
+        do_resize=True,
+        size=None,
+        do_normalize=True,
+        image_mean=[0.48145466, 0.4578275, 0.40821073],
+        image_std=[0.26862954, 0.26130258, 0.27577711],
+    ):
+        super().__init__()
+        size = size if size is not None else {"height": 1024, "width": 1024}
+        self.parent = parent
+        self.batch_size = batch_size
+        self.num_channels = num_channels
+        self.image_size = image_size
+        self.min_resolution = min_resolution
+        self.max_resolution = max_resolution
+        self.do_resize = do_resize
+        self.size = size
+        self.do_normalize = do_normalize
+        self.image_mean = image_mean
+        self.image_std = image_std
+
+    def prepare_image_processor_dict(self):
+        return {
+            "do_resize": self.do_resize,
+            "size": self.size,
+            "do_normalize": self.do_normalize,
+            "image_mean": self.image_mean,
+            "image_std": self.image_std,
+        }
+
+    def expected_output_image_shape(self, images):
+        return self.num_channels, self.size["height"], self.size["width"]
+
+    def prepare_image_inputs(self, equal_resolution=False, numpify=False, torchify=False):
+        return prepare_image_inputs(
+            batch_size=self.batch_size,
+            num_channels=self.num_channels,
+            min_resolution=self.min_resolution,
+            max_resolution=self.max_resolution,
+            equal_resolution=equal_resolution,
+            numpify=numpify,
+            torchify=torchify,
+        )
+
+
+@require_torch
+@require_vision
+class PPChart2TableImageProcessingTest(ImageProcessingTestMixin, unittest.TestCase):
+    def setUp(self):
+        super().setUp()
+        self.image_processor_tester = PPChart2TableImageProcessingTester(self)
+
+    @property
+    def image_processor_dict(self):
+        return self.image_processor_tester.prepare_image_processor_dict()
+
+    def test_image_processor_properties(self):
+        for image_processing_class in self.image_processing_classes.values():
+            image_processor = image_processing_class(**self.image_processor_dict)
+            self.assertTrue(hasattr(image_processor, "do_resize"))
+            self.assertTrue(hasattr(image_processor, "size"))
+            self.assertTrue(hasattr(image_processor, "do_normalize"))
+            self.assertTrue(hasattr(image_processor, "image_mean"))
+            self.assertTrue(hasattr(image_processor, "image_std"))

tests/models/pp_chart2table/test_modeling_pp_chart2table.py

@@ -0,0 +1,85 @@
+# 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.
+"""Testing suite for the PPChart2Table model."""
+
+import unittest
+
+from transformers import AutoModelForImageTextToText, AutoProcessor
+from transformers.testing_utils import cleanup, require_torch, require_vision, slow, torch_device
+
+from ...test_processing_common import url_to_local_path
+
+
+@slow
+@require_vision
+@require_torch
+class PPChart2TableIntegrationTest(unittest.TestCase):
+    def setUp(self):
+        model_path = "PaddlePaddle/PP-Chart2Table_safetensors"
+        self.model = AutoModelForImageTextToText.from_pretrained(model_path).to(torch_device)
+        self.processor = AutoProcessor.from_pretrained(model_path)
+        self.conversation = [
+            {
+                "role": "user",
+                "content": [
+                    {
+                        "type": "image",
+                        "url": url_to_local_path(
+                            "https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/chart_parsing_02.png"
+                        ),
+                    },
+                ],
+            },
+        ]
+
+    def tearDown(self):
+        cleanup(torch_device, gc_collect=True)
+
+    def test_small_model_integration_test_pp_chart2table(self):
+        inputs = self.processor.apply_chat_template(
+            self.conversation,
+            tokenize=True,
+            add_generation_prompt=True,
+            truncation=True,
+            return_dict=True,
+            return_tensors="pt",
+        ).to(self.model.device)
+
+        generated_ids = self.model.generate(**inputs, do_sample=False, max_new_tokens=32)
+        generated_ids_trimmed = [out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)]
+        decoded_output = self.processor.batch_decode(
+            generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
+        )
+
+        expected_output = ["年份 | 单家五星级旅游饭店年平均营收 (百万元) | 单家五星级旅游饭店年平均利润 (百万元)\n"]
+        self.assertEqual(decoded_output, expected_output)
+
+    def test_small_model_integration_test_pp_chart2table_batched(self):
+        inputs = self.processor.apply_chat_template(
+            [self.conversation, self.conversation],
+            tokenize=True,
+            add_generation_prompt=True,
+            truncation=True,
+            return_dict=True,
+            return_tensors="pt",
+        ).to(self.model.device)
+
+        generated_ids = self.model.generate(**inputs, do_sample=False, max_new_tokens=6)
+        generated_ids_trimmed = [out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)]
+        decoded_output = self.processor.batch_decode(
+            generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
+        )
+
+        expected_output = ["年份 | 单家", "年份 | 单家"]
+        self.assertEqual(decoded_output, expected_output)

tests/models/pp_chart2table/test_processing_pp_chart2table.py

@@ -0,0 +1,84 @@
+# Copyright 2026 The HuggingFace 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 transformers import PPChart2TableProcessor
+from transformers.testing_utils import require_vision
+
+from ...test_processing_common import ProcessorTesterMixin
+
+
+@require_vision
+class PPChart2TableProcessorTest(ProcessorTesterMixin, unittest.TestCase):
+    processor_class = PPChart2TableProcessor
+
+    @classmethod
+    def _setup_tokenizer(cls):
+        tokenizer_class = cls._get_component_class_from_processor("tokenizer")
+        tokenizer = tokenizer_class.from_pretrained("PaddlePaddle/PP-Chart2Table_safetensors")
+        return tokenizer
+
+    def test_ocr_queries(self):
+        processor = self.get_processor()
+        image_input = self.prepare_image_inputs()
+        conversation = [{"role": "user", "content": []}]
+        inputs = processor.apply_chat_template(
+            conversation,
+            tokenize=False,
+            add_generation_prompt=True,
+        )
+        inputs = processor(images=image_input, text=inputs, return_tensors="pt")
+        self.assertEqual(inputs["input_ids"].shape, (1, 286))
+        self.assertEqual(inputs["pixel_values"].shape, (1, 3, 1024, 1024))
+
+    def test_unstructured_kwargs_batched(self):
+        if "image_processor" not in self.processor_class.get_attributes():
+            self.skipTest(f"image_processor attribute not present in {self.processor_class}")
+        processor_components = self.prepare_components()
+        processor_kwargs = self.prepare_processor_dict()
+        processor = self.processor_class(**processor_components, **processor_kwargs)
+        self.skip_processor_without_typed_kwargs(processor)
+
+        input_str = self.prepare_text_inputs(batch_size=2, modalities="image")
+        image_input = self.prepare_image_inputs(batch_size=2)
+        inputs = processor(
+            text=input_str,
+            images=image_input,
+            return_tensors="pt",
+            do_rescale=True,
+            rescale_factor=-1.0,
+            padding="longest",
+            max_length=self.image_unstructured_max_length,
+        )
+
+        self.assertLessEqual(inputs[self.images_input_name][0][0].mean(), 0)
+
+    @unittest.skip(
+        reason="PPChart2Table relies on a heavily predetermined input format; chat template usage is not intended as expected"
+    )
+    def test_apply_chat_template_assistant_mask(self):
+        pass
+
+    @unittest.skip(
+        reason="PPChart2Table relies on a heavily predetermined input format; chat template usage is not intended as expected"
+    )
+    def test_apply_chat_template_image_0(self):
+        pass
+
+    @unittest.skip(
+        reason="PPChart2Table relies on a heavily predetermined input format; chat template usage is not intended as expected"
+    )
+    def test_apply_chat_template_image_1(self):
+        pass