File size: 1,391 Bytes
2ad713c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 |
#!/usr/bin/env python3
import json
import os
# Load the existing train.json file
with open('default/train.json', 'r') as f:
data = json.load(f)
# Create the directory for the new format
os.makedirs('default-fixed', exist_ok=True)
# Convert to the correct format (JSONL - one JSON object per line)
with open('default-fixed/train.jsonl', 'w') as f:
for i in range(len(data['file_name'])):
example = {
'file_name': data['file_name'][i],
'file_path': data['file_path'][i],
'file_size': data['file_size'][i],
'content_type': data['content_type'][i]
}
f.write(json.dumps(example) + '\n')
print(f"Converted {len(data['file_name'])} examples to JSONL format")
# Update the dataset_infos.json file
with open('dataset_infos.json', 'r') as f:
dataset_infos = json.load(f)
# Rename the default config to default-fixed
dataset_infos['default-fixed'] = dataset_infos.pop('default')
with open('dataset_infos.json', 'w') as f:
json.dump(dataset_infos, f, indent=2)
# Update the dataset_dict.json file
with open('dataset_dict.json', 'r') as f:
dataset_dict = json.load(f)
# Rename the default config to default-fixed
dataset_dict['default-fixed'] = dataset_dict.pop('default')
with open('dataset_dict.json', 'w') as f:
json.dump(dataset_dict, f, indent=2)
print("Updated dataset configuration files")
|