import json import os this_dir = os.path.dirname(os.path.abspath(__file__)) def mkdir(path): if not os.path.isdir(path): os.mkdir(path) mkdir(this_dir + '/../../data') mkdir(this_dir + '/../../data/helpsteer') for filename in ['train.jsonl', 'validation.jsonl']: with open(this_dir + '/' + filename, 'r') as f: lines = f.readlines() role_dict = {'prompt': 'user', 'response': 'assistant'} lines = [json.loads(line) for line in lines] conversations = [{'messages': [{'role': 'user', 'content': line['prompt']}, {'role': 'assistant', 'content': line['response']}]} for line in lines] print(conversations[0]) with open(this_dir + '/../../data/helpsteer/helpsteer_' + filename[0:-6] + '_all.jsonl', 'w') as f: f.writelines([json.dumps(conv) + '\n' for conv in conversations])