Datasets:
Tasks:
Question Answering
Modalities:
Text
Sub-tasks:
extractive-qa
Languages:
English
Size:
10K - 100K
ArXiv:
License:
| import pandas as pd | |
| import os | |
| from tqdm import tqdm | |
| import json | |
| import shutil | |
| directories = ['lifestyle', 'pooled', 'recreation', 'science', 'technology', 'writing'] #os.listdir() | |
| for directory in directories: | |
| print("Starting " + directory) | |
| for file_type in ["dev", "test"]: | |
| for question_type in ["qas.forum.jsonl", "qas.search.jsonl"]: | |
| if directory != 'pooled': | |
| with open('data/' + directory + "/" + file_type + "/metadata.jsonl", 'r', encoding="utf-8") as json_file: | |
| metadata = list(json_file) | |
| question_id_to_author = {} | |
| for json_str in metadata: | |
| current_row = json.loads(json_str) | |
| question_id_to_author[current_row["question_id"]] = current_row["question_author"] | |
| else: | |
| question_id_to_author = {} | |
| ######################################################## | |
| with open(directory + "/" + file_type + "_" + question_type, 'r', encoding="utf-8") as json_file: | |
| json_list = list(json_file) | |
| revised_jsonl = [] | |
| for json_str in tqdm(json_list): | |
| current_row = json.loads(json_str) | |
| try: | |
| current_row["question_author"] = question_id_to_author[current_row['qid']] | |
| except: | |
| current_row["question_author"] = "" | |
| revised_jsonl.append(current_row) | |
| ######################################################## | |
| if not os.path.isdir(directory): | |
| os.mkdir(directory) | |
| with open(directory + "/" + file_type + "_" + question_type, 'w', encoding="utf-8") as outfile: | |
| for entry in revised_jsonl: | |
| json.dump(entry, outfile) | |
| outfile.write('\n') | |
| #with open(directory + "/" + file_type + question_type, 'w', encoding="utf-8") as fout: | |
| # json.dump(data, fout) | |
| #shutil.move('data/' + directory + "/" + file_type + "/" + question_type, | |
| # directory + "/" + file_type + "_" + question_type) |