[update]add code

.gitattributes

@@ -53,3 +53,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.jpg filter=lfs diff=lfs merge=lfs -text
 *.jpeg filter=lfs diff=lfs merge=lfs -text
 *.webp filter=lfs diff=lfs merge=lfs -text
+*.jsonl filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,12 @@
+
+.git/
+.idea/
+
+data/
+examples/preprocess/data/
+hub_datasets/
+
+**/__pycache__/
+
+**/*.csv
+**/*.xlsx

README.md

@@ -11,30 +11,24 @@ size_categories:
 ## 句子对数据集
 
 数据集从网上收集整理如下:
-| 数据 | 原始数据/项目地址 | 样本个数 | 原始数据描述 | 替代数据下载地址 |
-| :--- |     :---:      |   :---: |    :---:    |     :---:     |
-| 数据 | 原始数据/项目地址 | 样本个数 | 原始数据描述 | 替代数据下载地址 |
+| 数据 | 语言 | 原始数据/项目地址 | 样本个数 | 原始数据描述 | 替代数据下载地址 |
+| :--- | :---: |     :---:      |   :---: |    :---:    |     :---:     |
+| ChineseSTS | 汉语 | [ChineseSTS](https://github.com/IAdmireu/ChineseSTS) | 24.7K | STS 中文文本语义相似度（使用时注意打乱数据集） | [ChineseSTS](https://huggingface.co/datasets/tiansz/ChineseSTS) |
+| ccks2018_task3 | 汉语 | [CCKS2018_3](https://www.biendata.xyz/competition/CCKS2018_3/data/) | 100K | CCKS 2018 微众银行智能客服问句匹配大赛 |  |
+| DIAC2019 | 汉语 | [DIAC2019](https://www.biendata.xyz/competition/2019diac/data/) | 6K | 以问题组的形式提供，每组问句又分为等价部分和不等价部分，等价问句之间互相组合可以生成正样本，等价问句和不等价问句之间互相组合可以生成负样本。我们提供6000组问句的训练集。 |  |
+| LCQMC | 汉语 | [LCQMC](https://www.luge.ai/#/luge/dataDetail?id=14); [C18-1166.pdf](https://aclanthology.org/C18-1166.pdf) | TRAIN: 238766, VALID: 8802, TEST: 12500 | 百度知道领域的中文问题匹配数据集，目的是为了解决在中文领域大规模问题匹配数据集的缺失。该数据集从百度知道不同领域的用户问题中抽取构建数据。| [lcqmc_data](https://github.com/xiaohai-AI/lcqmc_data) |
+| AFQMC | 汉语 | [AFQMC](https://tianchi.aliyun.com/dataset/106411) | TRAIN: 34334, VALID: 4316, TEST: 3861 | 蚂蚁金融语义相似度数据集，用于问题相似度计算。即：给定客服里用户描述的两句话，用算法来判断是否表示了相同的语义。 |  |
+
 
 <details>
 <summary>参考的数据来源,展开查看</summary>
 <pre><code>
 
-https://huggingface.co/datasets/fathyshalab/atis_intents
-https://huggingface.co/datasets/generalization/conv_intent_Full-p_1
-https://huggingface.co/datasets/banking77
-https://huggingface.co/datasets/dipesh/Intent-Classification-large
-https://huggingface.co/datasets/SetFit/amazon_massive_intent_en-US
-https://huggingface.co/datasets/SetFit/amazon_massive_intent_zh-CN
-https://huggingface.co/datasets/SetFit/amazon_massive_intent_zh-TW
-https://huggingface.co/datasets/snips_built_in_intents
-https://huggingface.co/datasets/zapsdcn/citation_intent
-https://huggingface.co/datasets/ibm/vira-intents
-https://huggingface.co/datasets/mteb/mtop_intent
-https://huggingface.co/datasets/Bhuvaneshwari/intent_classification
-https://huggingface.co/datasets/ibm/vira-intents-live
-https://huggingface.co/datasets/ebrigham/nl_banking_intents
-https://pan.baidu.com/s/19_oqY4bC_lJa_7Mc6lxU7w?pwd=v4bi
-https://gitee.com/a2798063/SMP2019/tree/master
+https://github.com/liucongg/NLPDataSet
+
+https://huggingface.co/datasets/tiansz/ChineseSTS
+https://zhuanlan.zhihu.com/p/454173790
+
 
 </code></pre>
 </details>

examples/preprocess/process_afqmc.py

@@ -0,0 +1,80 @@
+#!/usr/bin/python3
+# -*- coding: utf-8 -*-
+import argparse
+import json
+import os
+from pathlib import Path
+import sys
+
+pwd = os.path.abspath(os.path.dirname(__file__))
+sys.path.append(os.path.join(pwd, '../../'))
+
+from datasets import load_dataset
+from lxml import etree
+import pandas as pd
+from tqdm import tqdm
+
+from project_settings import project_path
+
+
+def get_args():
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument("--data_dir", default="./data/afqmc", type=str)
+    parser.add_argument(
+        "--output_file",
+        default=(project_path / "data/afqmc.jsonl"),
+        type=str
+    )
+
+    args = parser.parse_args()
+    return args
+
+
+def main():
+    args = get_args()
+
+    data_dir = Path(args.data_dir)
+
+    with open(args.output_file, "w", encoding="utf-8") as fout:
+        for name in ["train.json", "dev.json", "test.json"]:
+            filename = data_dir / name
+
+            with open(filename.as_posix(), "r", encoding="utf-8") as fin:
+                for row in fin:
+                    row = json.loads(row)
+
+                    sentence1 = row["sentence1"]
+                    sentence2 = row["sentence2"]
+
+                    if name == "train.json":
+                        label = row["label"]
+                        flag = "train"
+                    elif name == "dev.json":
+                        label = row["label"]
+                        flag = "validation"
+                    elif name == "test.json":
+                        label = None
+                        flag = "test"
+                    else:
+                        raise AssertionError
+
+                    if label not in ("0", "1", None):
+                        raise AssertionError
+
+                    row = {
+                        "sentence1": sentence1,
+                        "sentence2": sentence2,
+                        "label": label,
+                        "data_source": "lcqmc",
+                        "split": flag
+                    }
+
+                    row = json.dumps(row, ensure_ascii=False)
+                    fout.write("{}\n".format(row))
+
+    return
+
+
+if __name__ == '__main__':
+    main()

examples/preprocess/process_ccks2018_task3.py

@@ -0,0 +1,79 @@
+#!/usr/bin/python3
+# -*- coding: utf-8 -*-
+import argparse
+import json
+import os
+from pathlib import Path
+import sys
+
+pwd = os.path.abspath(os.path.dirname(__file__))
+sys.path.append(os.path.join(pwd, '../../'))
+
+from datasets import load_dataset
+from tqdm import tqdm
+
+from project_settings import project_path
+
+
+def get_args():
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument("--data_dir", default="./data/ccks2018_task3", type=str)
+    parser.add_argument(
+        "--output_file",
+        default=(project_path / "data/ccks2018_task3.jsonl"),
+        type=str
+    )
+
+    args = parser.parse_args()
+    return args
+
+
+def main():
+    args = get_args()
+
+    data_dir = Path(args.data_dir)
+
+    with open(args.output_file, "w", encoding="utf-8") as f:
+        for name in ["task3_train.txt", "task3_dev.txt", "test_with_id.txt"]:
+            filename = data_dir / name
+            with open(filename, "r", encoding="utf-8") as fin:
+                for row in fin:
+                    row = str(row).strip()
+                    splits = row.split("\t")
+                    # print(splits)
+                    if name == "task3_train.txt":
+                        sentence1, sentence2, label = splits
+                        flag = "train"
+                    elif name == "task3_dev.txt":
+                        sentence1, sentence2 = splits[-2:]
+                        label = "1"
+                        flag = "validation"
+                    elif name == "test_with_id.txt":
+                        sentence1, sentence2 = splits[-2:]
+                        label = None
+                        flag = "test"
+                    else:
+                        raise AssertionError
+
+                    label = str(int(label)) if label is not None else None
+
+                    if label not in ("0", "1", None):
+                        raise AssertionError
+
+                    row = {
+                        "sentence1": sentence1,
+                        "sentence2": sentence2,
+                        "label": label,
+                        "data_source": "ccks2018_task3",
+                        "split": flag
+                    }
+
+                    row = json.dumps(row, ensure_ascii=False)
+                    f.write("{}\n".format(row))
+
+    return
+
+
+if __name__ == '__main__':
+    main()

examples/preprocess/process_chinese_sts.py

@@ -0,0 +1,78 @@
+#!/usr/bin/python3
+# -*- coding: utf-8 -*-
+import argparse
+import json
+import os
+import sys
+
+pwd = os.path.abspath(os.path.dirname(__file__))
+sys.path.append(os.path.join(pwd, '../../'))
+
+from datasets import load_dataset
+from tqdm import tqdm
+
+from project_settings import project_path
+
+
+def get_args():
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument("--dataset_path", default="tiansz/ChineseSTS", type=str)
+    parser.add_argument(
+        "--dataset_cache_dir",
+        default=(project_path / "hub_datasets").as_posix(),
+        type=str
+    )
+    parser.add_argument(
+        "--output_file",
+        default=(project_path / "data/chinese_sts.jsonl"),
+        type=str
+    )
+
+    args = parser.parse_args()
+    return args
+
+
+def main():
+    args = get_args()
+
+    dataset_dict = load_dataset(
+        path=args.dataset_path,
+        cache_dir=args.dataset_cache_dir,
+    )
+    print(dataset_dict)
+    with open(args.output_file, "w", encoding="utf-8") as f:
+        for sample in tqdm(dataset_dict["train"]):
+            text = sample["text"]
+
+            if text == "句子1\t句子2\t相似度":
+                continue
+
+            splits = text.split("\t")
+            if len(splits) != 3:
+                raise AssertionError
+
+            sentence1 = splits[0]
+            sentence2 = splits[1]
+            label = splits[2]
+            label = str(int(float(label)))
+
+            if label not in ("0", "1", None):
+                raise AssertionError
+
+            row = {
+                "sentence1": sentence1,
+                "sentence2": sentence2,
+                "label": label,
+                "data_source": "ChineseSTS",
+                "split": "train"
+            }
+
+            row = json.dumps(row, ensure_ascii=False)
+            f.write("{}\n".format(row))
+
+    return
+
+
+if __name__ == '__main__':
+    main()

examples/preprocess/process_diac2019.py

@@ -0,0 +1,128 @@
+#!/usr/bin/python3
+# -*- coding: utf-8 -*-
+import argparse
+import json
+import os
+from pathlib import Path
+import sys
+
+pwd = os.path.abspath(os.path.dirname(__file__))
+sys.path.append(os.path.join(pwd, '../../'))
+
+from datasets import load_dataset
+from lxml import etree
+import pandas as pd
+from tqdm import tqdm
+
+from project_settings import project_path
+
+
+def get_args():
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument("--data_dir", default="./data/diac2019", type=str)
+    parser.add_argument(
+        "--output_file",
+        default=(project_path / "data/diac2019.jsonl"),
+        type=str
+    )
+
+    args = parser.parse_args()
+    return args
+
+
+def main():
+    args = get_args()
+
+    data_dir = Path(args.data_dir)
+
+    with open(args.output_file, "w", encoding="utf-8") as f:
+        # train_set.xml
+        filename = data_dir / "train_set.xml"
+        tree = etree.parse(filename.as_posix())
+        root = tree.getroot()
+
+        for group in root:
+            equal = group[0]
+            not_equal = group[1]
+
+            equal_questions = [question.text for question in equal]
+            not_equal_questions = [question.text for question in not_equal]
+
+            pairs = set()
+
+            for q1 in equal_questions:
+                for q2 in equal_questions:
+                    if q1 == q2:
+                        continue
+                    pair = (q1, q2, 1)
+                    pairs.add(pair)
+
+            for q1 in equal_questions:
+                for q2 in not_equal_questions:
+                    if q1 == q2:
+                        continue
+                    pair = (q1, q2, 0)
+                    pairs.add(pair)
+
+            for pair in pairs:
+                q1, q2, label = pair
+
+                label = str(label)
+                if label not in ("0", "1", None):
+                    raise AssertionError
+
+                json_row = {
+                    "sentence1": q1,
+                    "sentence2": q2,
+                    "label": label,
+                    "data_source": "diac2019",
+                    "split": "train"
+                }
+
+                json_row = json.dumps(json_row, ensure_ascii=False)
+                f.write("{}\n".format(json_row))
+
+        # dev_set.csv
+        filename = data_dir / "dev_set.csv"
+        df = pd.read_csv(filename.as_posix(), delimiter="\t")
+        for i, row in df.iterrows():
+
+            question1 = row["question1"]
+            question2 = row["question2"]
+
+            json_row = {
+                "sentence1": question1,
+                "sentence2": question2,
+                "score": "1",
+                "data_source": "diac2019",
+                "split": "validation"
+            }
+
+            json_row = json.dumps(json_row, ensure_ascii=False)
+            f.write("{}\n".format(json_row))
+
+        # test_set.csv
+        filename = data_dir / "test_set.csv"
+        df = pd.read_csv(filename.as_posix(), delimiter="\t")
+        for i, row in df.iterrows():
+
+            question1 = row["question1"]
+            question2 = row["question2"]
+
+            json_row = {
+                "sentence1": question1,
+                "sentence2": question2,
+                "score": None,
+                "data_source": "diac2019",
+                "split": "test"
+            }
+
+            json_row = json.dumps(json_row, ensure_ascii=False)
+            f.write("{}\n".format(json_row))
+
+    return
+
+
+if __name__ == '__main__':
+    main()

examples/preprocess/process_lcqmc.py

@@ -0,0 +1,81 @@
+#!/usr/bin/python3
+# -*- coding: utf-8 -*-
+import argparse
+import json
+import os
+from pathlib import Path
+import sys
+
+pwd = os.path.abspath(os.path.dirname(__file__))
+sys.path.append(os.path.join(pwd, '../../'))
+
+from datasets import load_dataset
+from lxml import etree
+import pandas as pd
+from tqdm import tqdm
+
+from project_settings import project_path
+
+
+def get_args():
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument("--data_dir", default="./data/lcqmc", type=str)
+    parser.add_argument(
+        "--output_file",
+        default=(project_path / "data/lcqmc.jsonl"),
+        type=str
+    )
+
+    args = parser.parse_args()
+    return args
+
+
+def main():
+    args = get_args()
+
+    data_dir = Path(args.data_dir)
+
+    with open(args.output_file, "w", encoding="utf-8") as fout:
+        for name in ["train.tsv", "dev.tsv", "test.tsv"]:
+            filename = data_dir / name
+
+            with open(filename.as_posix(), "r", encoding="utf-8") as fin:
+                for row in fin:
+                    row = str(row).strip()
+                    splits = row.split("\t")
+
+                    if name == "train.tsv":
+                        q1, q2, label = splits
+                        flag = "train"
+                    elif name == "dev.tsv":
+                        q1, q2, label = splits
+                        flag = "validation"
+                    elif name == "test.tsv":
+                        q1, q2 = splits
+                        label = None
+                        flag = "test"
+                    else:
+                        raise AssertionError
+
+                    label = str(int(label)) if label is not None else None
+
+                    if label not in ("0", "1", None):
+                        raise AssertionError
+
+                    row = {
+                        "sentence1": q1,
+                        "sentence2": q2,
+                        "label": label,
+                        "data_source": "lcqmc",
+                        "split": flag
+                    }
+
+                    row = json.dumps(row, ensure_ascii=False)
+                    fout.write("{}\n".format(row))
+
+    return
+
+
+if __name__ == '__main__':
+    main()

main.py

@@ -0,0 +1,16 @@
+#!/usr/bin/python3
+# -*- coding: utf-8 -*-
+from datasets import load_dataset
+
+dataset = load_dataset(
+    "sentence_pair.py",
+    name="ccks2018_task3",
+    split="validation",
+)
+
+for sample in dataset:
+    print(sample)
+
+
+if __name__ == '__main__':
+    pass

project_settings.py

@@ -0,0 +1,12 @@
+#!/usr/bin/python3
+# -*- coding: utf-8 -*-
+import os
+from pathlib import Path
+
+
+project_path = os.path.abspath(os.path.dirname(__file__))
+project_path = Path(project_path)
+
+
+if __name__ == '__main__':
+    pass

requirements.txt

@@ -0,0 +1,7 @@
+datasets
+tqdm==4.66.1
+pandas==2.0.3
+xlrd==1.2.0
+openpyxl==3.0.9
+fsspec==2023.9.2
+lxml==4.9.3

sentence_pair.py

@@ -0,0 +1,107 @@
+#!/usr/bin/python3
+# -*- coding: utf-8 -*-
+from glob import glob
+import json
+import os
+from pathlib import Path
+
+import datasets
+
+
+_URLS = {
+    "ccks2018_task3": "data/ccks2018_task3.jsonl",
+    "chinese_sts": "data/chinese_sts.jsonl",
+    "diac2019": "data/diac2019.jsonl",
+    "lcqmc": "data/lcqmc.jsonl",
+
+}
+
+
+_CITATION = """\
+@dataset{sentence_pair,
+  author       = {Xing Tian},
+  title        = {sentence_pair},
+  month        = sep,
+  year         = 2023,
+  publisher    = {Xing Tian},
+  version      = {1.0},
+}
+"""
+
+
+class SentencePair(datasets.GeneratorBasedBuilder):
+    VERSION = datasets.Version("1.0.0")
+
+    configs = list()
+    for name in _URLS.keys():
+        config = datasets.BuilderConfig(name=name, version=VERSION, description=name)
+        configs.append(config)
+
+    BUILDER_CONFIGS = [
+        *configs
+    ]
+
+    def _info(self):
+        features = datasets.Features(
+            {
+                "sentence1": datasets.Value("string"),
+                "sentence2": datasets.Value("string"),
+                "label": datasets.Value("string"),
+                "data_source": datasets.Value("string"),
+                "split": datasets.Value("string"),
+            }
+        )
+        return datasets.DatasetInfo(
+            features=features,
+            supervised_keys=None,
+            homepage="",
+            license="",
+            citation=_CITATION,
+        )
+
+    def _split_generators(self, dl_manager):
+        """Returns SplitGenerators."""
+        url = _URLS[self.config.name]
+        dl_path = dl_manager.download(url)
+        archive_path = dl_path
+
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={"archive_path": archive_path, "split": "train"},
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.VALIDATION,
+                gen_kwargs={"archive_path": archive_path, "split": "validation"},
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.TEST,
+                gen_kwargs={"archive_path": archive_path, "split": "test"},
+            ),
+        ]
+
+    def _generate_examples(self, archive_path, split):
+        """Yields examples."""
+        archive_path = Path(archive_path)
+
+        idx = 0
+
+        with open(archive_path, "r", encoding="utf-8") as f:
+            for row in f:
+                sample = json.loads(row)
+
+                if sample["split"] != split:
+                    continue
+
+                yield idx, {
+                    "sentence1": sample["sentence1"],
+                    "sentence2": sample["sentence2"],
+                    "label": sample["label"],
+                    "data_source": sample["data_source"],
+                    "split": sample["split"],
+                }
+                idx += 1
+
+
+if __name__ == '__main__':
+    pass