Upload LILT2.py

LILT2.py

@@ -0,0 +1,191 @@
+import json
+import os
+
+from PIL import Image
+
+import datasets
+from datasets import load_dataset
+
+def load_image(image_path):
+    image = Image.open(image_path).convert("RGB")
+    w, h = image.size
+    return image, (w, h)
+
+def normalize_bbox(bbox, size):
+    return [
+        int(1000 * bbox[0] / size[0]),
+        int(1000 * bbox[1] / size[1]),
+        int(1000 * bbox[2] / size[0]),
+        int(1000 * bbox[3] / size[1]),
+    ]
+
+logger = datasets.logging.get_logger(__name__)
+
+
+_CITATION = """\
+
+}
+"""
+
+_DESCRIPTION = """\
+
+"""
+
+
+class FunsdConfig(datasets.BuilderConfig):
+    """BuilderConfig for FUNSD"""
+
+    def __init__(self, **kwargs):
+        """BuilderConfig for FUNSD.
+        Args:
+          **kwargs: keyword arguments forwarded to super.
+        """
+        super(FunsdConfig, self).__init__(**kwargs)
+
+
+class Funsd(datasets.GeneratorBasedBuilder):
+    """Conll2003 dataset."""
+
+    BUILDER_CONFIGS = [
+        FunsdConfig(name="funsd", version=datasets.Version("1.0.0"), description="FUNSD dataset"),
+    ]
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=datasets.Features(
+                {
+                    "id": datasets.Value("string"),
+                    "tokens": datasets.Sequence(datasets.Value("string")),
+                    "bboxes": datasets.Sequence(datasets.Sequence(datasets.Value("int64"))),
+                    "ner_tags": datasets.Sequence(
+                        datasets.features.ClassLabel(
+                            names = [           "S-ANSWER_EXP",
+                                                "S-ANSWER_FECHA_SERVICIO",
+                                                "S-ANSWER_HORA_SERVICIO",
+                                                "S-ANSWER_SALA",
+                                                "B-ANSWER_NOMBRE1",
+                                                "I-ANSWER_NOMBRE1",
+                                                "E-ANSWER_NOMBRE1",
+                                                "B-ANSWER_DIRECCION",
+                                                "I-ANSWER_DIRECCION",
+                                                "E-ANSWER_DIRECCION",
+                                                "B-ANSWER_POBLACION",
+                                                "I-ANSWER_POBLACION",
+                                                "E-ANSWER_POBLACION",
+                                                "S-ANSWER_DNI",
+                                                "S-ANSWER_TELEFONO",
+                                                "S-ANSWER_EDAD",
+                                                "S-ANSWER_NACIMIENTO_DIF",
+                                                "S-ANSWER_ESTADO_CIVIL_DIF",
+                                                "S-ANSWER_FECHA_DEF",
+                                                "B-ANSWER_LUGAR_DEF",
+                                                "I-ANSWER_LUGAR_DEF",
+                                                "E-ANSWER_LUGAR_DEF",
+                                                "S-ANSWER_NATURAL_DE_DIF",
+                                                "B-ANSWER_PADRES_DIF",
+                                                "I-ANSWER_PADRES_DIF",
+                                                "E-ANSWER_PADRES_DIF",
+                                                "B-ANSWER_NOMBRE_TITULAR",
+                                                "I-ANSWER_NOMBRE_TITULAR",
+                                                "E-ANSWER_NOMBRE_TITULAR",
+                                                "S-ANSWER_AUT_DNI_TITULAR",
+                                                "B-ANSWER_DIRECCION_TITULAR",
+                                                "I-ANSWER_DIRECCION_TITULAR",
+                                                "E-ANSWER_DIRECCION_TITULAR",
+                                                "B-ANSWER_POBLACION_TITULAR",
+                                                "I-ANSWER_POBLACION_TITULAR",
+                                                "E-ANSWER_POBLACION_TITULAR",
+                                                "B-ANSWER_AUTORIZACION_TITULAR",
+                                                "I-ANSWER_AUTORIZACION_TITULAR",
+                                                "E-ANSWER_AUTORIZACION_TITULAR",
+                                                "S-ANSWER_DNI_TITULAR",
+                                                "S-ANSWER_HORA_DEFUNCION",
+                                                "B-ANSWER_DESCRIPCION",
+                                                "I-ANSWER_DESCRIPCION",
+                                                "E-ANSWER_DESCRIPCION",
+                                                "B-ANSWER_NOMBRE",
+                                                "I-ANSWER_NOMBRE",
+                                                "E-ANSWER_NOMBRE",
+                                                "S-ANSWER_CANTIDAD",
+                                                "S-ANSWER_IMPORTE"]
+                        )
+                    ),
+                    "image": datasets.features.Image(),
+                }
+            ),
+            supervised_keys=None,
+            homepage="",
+            citation=_CITATION,
+        )
+
+    def _split_generators(self, dl_manager):
+        """Returns SplitGenerators."""
+        downloaded_file = dl_manager.download_and_extract("https://huggingface.co/datasets/LauraExp/LILT2/resolve/main/Data.zip")
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN, gen_kwargs={"filepath": f"{downloaded_file}/Data/training_data/"}
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.TEST, gen_kwargs={"filepath": f"{downloaded_file}/Data/testing_data/"}
+            ),
+        ]
+
+    def get_line_bbox(self, bboxs):
+        x = [bboxs[i][j] for i in range(len(bboxs)) for j in range(0, len(bboxs[i]), 2)]
+        y = [bboxs[i][j] for i in range(len(bboxs)) for j in range(1, len(bboxs[i]), 2)]
+
+        x0, y0, x1, y1 = min(x), min(y), max(x), max(y)
+
+        assert x1 >= x0 and y1 >= y0
+        bbox = [[x0, y0, x1, y1] for _ in range(len(bboxs))]
+        return bbox
+
+    def _generate_examples(self, filepath):
+        logger.info("⏳ Generating examples from = %s", filepath)
+        ann_dir = os.path.join(filepath, "annotations")
+        img_dir = os.path.join(filepath, "images")
+        for guid, file in enumerate(sorted(os.listdir(ann_dir))):
+            tokens = []
+            bboxes = []
+            ner_tags = []
+
+            file_path = os.path.join(ann_dir, file)
+            with open(file_path, "r", encoding="utf8") as f:
+                data = json.load(f)
+            image_path = os.path.join(img_dir, file)
+            image_path = image_path.replace("json", "png")
+            image, size = load_image(image_path)
+            for item in data["form"]:
+                words_example, label = item["words"], item["label"]
+                words_example = [w for w in words_example if w["text"].strip() != ""]
+                if len(words_example) == 0:
+                    continue
+                if label == "other":
+                    for w in words_example:
+                        tokens.append(w["text"])
+                        ner_tags.append("O")
+                        bboxes.append(normalize_bbox(w["box"], size))
+                else:
+                    if len(words_example) == 1:
+                        tokens.append(words_example[0]["text"])
+                        ner_tags.append("S-" + label.upper())
+                        bboxes.append(normalize_bbox(words_example[0]["box"], size))
+                    else:
+                        tokens.append(words_example[0]["text"])
+                        ner_tags.append("B-" + label.upper())
+                        bboxes.append(normalize_bbox(words_example[0]["box"], size))
+                        for w in words_example[1:]:
+                            tokens.append(w["text"])
+                            ner_tags.append("I-" + label.upper())
+                            bboxes.append(normalize_bbox(w["box"], size))
+                        tokens.append(words_example[-1]["text"])
+                        ner_tags.append("E-" + label.upper())
+                        bboxes.append(normalize_bbox(words_example[-1]["box"], size))
+                        cur_line_bboxes = self.get_line_bbox(cur_line_bboxes)
+                        bboxes.extend(cur_line_bboxes)
+            yield guid, {"id": str(guid), "tokens": tokens, "bboxes": bboxes, "ner_tags": ner_tags,
+                         "image": image}
+
+
+