File size: 1,573 Bytes
c96d322 |
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 |
import json
def extract_unique_in_context_examples(input_file, output_file):
# Read the original JSON file
with open(input_file, 'r') as file:
data = json.load(file)
# Set for tracking unique examples and test questions
unique_examples = set()
test_questions = set()
# First, collect all test questions
for key, value in data.items():
if "test_example" in value:
test_question = json.dumps(value["test_example"]["question"], sort_keys=True)
test_questions.add(test_question)
# List to store unique in-context examples
example_pool = []
# Extract unique in-context examples
for key, value in data.items():
for example_key, example_value in value.items():
if 'in_context_example' in example_key:
# Convert the example to a JSON string for comparison
example_str = json.dumps(example_value["question"], sort_keys=True)
if example_str not in unique_examples and example_str not in test_questions:
unique_examples.add(example_str)
example_pool.append(example_value)
# Create the final dictionary with the example pool
print(len(example_pool))
final_data = {"example_pool": example_pool}
# Write the extracted data to a new JSON file
with open(output_file, 'w') as file:
json.dump(final_data, file, indent=4)
# Usage
input_file = '345hop_random_true.json'
output_file = 'example_pool.json'
extract_unique_in_context_examples(input_file, output_file)
|