| """PorSimplesSent dataset""" | |
| import datasets | |
| import pandas as pd | |
| _CITATION = """ | |
| """ | |
| _DESCRIPTION = """ | |
| """ | |
| _URLS = { | |
| "nat_str": "https://raw.githubusercontent.com/sidleal/porsimplessent/c0b7bb6ccda6b40ebd7f5524b08a5699b2266ffe/pss/pss2_align_length_nat_str.tsv", | |
| "ori_nat": "https://raw.githubusercontent.com/sidleal/porsimplessent/c0b7bb6ccda6b40ebd7f5524b08a5699b2266ffe/pss/pss2_align_length_ori_nat.tsv", | |
| "ori_str": "https://raw.githubusercontent.com/sidleal/porsimplessent/c0b7bb6ccda6b40ebd7f5524b08a5699b2266ffe/pss/pss2_align_length_ori_str.tsv" | |
| } | |
| class PorSimplesSent(datasets.GeneratorBasedBuilder): | |
| VERSION = datasets.Version("1.0.0") | |
| def _info(self): | |
| return datasets.DatasetInfo( | |
| description=_DESCRIPTION, | |
| features=datasets.Features( | |
| { | |
| "sentence1": datasets.Value("string"), | |
| "sentence2": datasets.Value("string"), | |
| "label": datasets.Value("int32"), | |
| "production_id": datasets.Value("int32"), | |
| "level": datasets.Value("string"), | |
| "changed": datasets.Value("string"), | |
| "split": datasets.Value("string"), | |
| "sentence_text_from": datasets.Value("string"), | |
| "sentence_text_to": datasets.Value("string"), | |
| }), | |
| supervised_keys=None, | |
| homepage="", | |
| citation=_CITATION, | |
| ) | |
| def _split_generators(self, dl_manager): | |
| downloaded_files = dl_manager.download(_URLS) | |
| return [ | |
| datasets.SplitGenerator( | |
| name=datasets.Split.TRAIN, | |
| gen_kwargs={ | |
| "nat_str": downloaded_files["nat_str"], | |
| "ori_nat": downloaded_files["ori_nat"], | |
| "ori_str": downloaded_files["ori_str"], | |
| "split": "train" | |
| } | |
| ), | |
| datasets.SplitGenerator( | |
| name=datasets.Split.VALIDATION, | |
| gen_kwargs={ | |
| "nat_str": downloaded_files["nat_str"], | |
| "ori_nat": downloaded_files["ori_nat"], | |
| "ori_str": downloaded_files["ori_str"], | |
| "split": "validation" | |
| } | |
| ), | |
| datasets.SplitGenerator( | |
| name=datasets.Split.TEST, | |
| gen_kwargs={ | |
| "nat_str": downloaded_files["nat_str"], | |
| "ori_nat": downloaded_files["ori_nat"], | |
| "ori_str": downloaded_files["ori_str"], | |
| "split": "test" | |
| } | |
| ) | |
| ] | |
| def _generate_examples(self, nat_str, ori_nat, ori_str, split): | |
| df1 = pd.read_csv(nat_str, sep="\t", on_bad_lines="skip") | |
| df2 = pd.read_csv(ori_nat, sep="\t", on_bad_lines="skip") | |
| df3 = pd.read_csv(ori_str, sep="\t", on_bad_lines="skip") | |
| df = pd.concat([df1, df2, df3], axis=0) | |
| df["mod"] = df["production_id"].apply(lambda x: x % 5) | |
| if split == "validation": | |
| df = df[df["mod"] == 1] | |
| elif split == "test": | |
| df = df[df["mod"] == 2] | |
| elif split == "train": | |
| df = df[df["mod"] != 1] | |
| df = df[df["mod"] != 2] | |
| df = df.sort_values(by=["production_id", "sentence_text_from", "sentence_text_to"]) | |
| prod_id_set = sorted(list(set(df["production_id"].values.tolist()))) | |
| records = df.to_dict("records") | |
| for idx, item in enumerate(records): | |
| row = {} | |
| for key in ["production_id", "level", "changed", "split", "sentence_text_from", "sentence_text_to"]: | |
| row[key] = item[key] | |
| if item["changed"] == "S": | |
| if prod_id_set.index(item["production_id"]) % 2 == 0: | |
| row["sentence1"] = item["sentence_text_from"] | |
| row["sentence2"] = item["sentence_text_to"] | |
| row["label"] = 2 | |
| else: | |
| row["sentence1"] = item["sentence_text_to"] | |
| row["sentence2"] = item["sentence_text_from"] | |
| row["label"] = 0 | |
| else: | |
| row["sentence1"] = item["sentence_text_from"] | |
| row["sentence2"] = item["sentence_text_to"] | |
| row["label"] = 1 | |
| yield idx, row |