Created
April 11, 2024 12:08
-
-
Save zhaoyanpeng/3ff976f6215b55cb4d6deb4725e9e218 to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| droot = "/home/xxxx/data" | |
| kroot = f"{droot}/kelm-corpus/updated-2021" | |
| kname = "kelm_generated_corpus.jsonl" | |
| ename = "entities.jsonl" | |
| tekgen_train = "quadruples-train.tsv" | |
| tekgen_test = "quadruples-test.tsv" | |
| tekgen_val = "quadruples-validation.tsv" | |
| tekgen_train_file = f"{kroot}/{tekgen_train}" | |
| tekgen_test_file = f"{kroot}/{tekgen_test}" | |
| tekgen_val_file = f"{kroot}/{tekgen_val}" | |
| kfile = f"{kroot}/{kname}" | |
| efile = f"{kroot}/{ename}" | |
| def serialize_triples(jsons, ofile=None): | |
| def replace_special(s): | |
| # reserved | |
| s = s.replace("{", "(") | |
| s = s.replace("}", ")") | |
| # reserved | |
| # s = s.replace("::", ":") | |
| return s | |
| def serialize_tri(d): | |
| r, o = [replace_special(x) for x in d[1:]] | |
| r = re.sub("[:]{2,}", ":", r) # appear 1 time at most | |
| return f"{{ {r}, {o} }}" | |
| def serialize_quad(d): | |
| o, r, v = [replace_special(x) for x in d[1:]] | |
| rv = f"{r}::{v}" | |
| rv = re.sub("[:]{3,}", "::", rv) # appear 2 times at most | |
| return f"{{ {rv}, {o} }}" | |
| debug = True | |
| max_line = 100 | |
| print_quad = True | |
| writer = jsonlines.open(ofile, "w") if ofile is not None else None | |
| # serization_all = list() | |
| sentence_key = (jsons.columns.values)[2] | |
| for i in tqdm(range(len(jsons))): | |
| row = jsons.iloc[i] | |
| triples = row["triples"] | |
| try: | |
| root = replace_special(triples[0][0]) | |
| children = list() | |
| for tp in triples: | |
| if len(tp) == 3: | |
| children.append(serialize_tri(tp)) | |
| elif len(tp) == 4: | |
| children.append(serialize_quad(tp)) | |
| print_quad = False | |
| except: | |
| continue | |
| children = "{ " + " ".join(children) + " }" | |
| serization = "{ " + root + " " + children + " }" | |
| # serization_all.append(serization) | |
| if writer is not None: | |
| writer.write({ | |
| "bracketed": serization, | |
| "sentence": row[sentence_key] | |
| }) | |
| if False and i >= max_line: | |
| break | |
| # if debug or not print_quad: | |
| # print(f"{row[0]}\n\n{row[1]}\n\n{row[2]}\n") | |
| # print(serization) | |
| # break | |
| if writer is not None: | |
| print(f"Saving to `{ofile}`") | |
| writer.close() | |
| data_file = kfile # tekgen_val_file # tekgen_test_file # tekgen_train_file # | |
| json_data = json_kelm # json_val # json_test # json_train # | |
| ofile = data_file.rsplit(".", 1)[0] + ".bracketed.jsonl" | |
| serialize_triples(json_data, ofile) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment