t5-nynorsk-norbench / corpus /calculate_bleu.py
pere's picture
calculations
645191f
raw
history blame
No virus
720 Bytes
import json
import sys
import sacrebleu
# Check command-line arguments
if len(sys.argv) != 2:
print("Usage: python script.py <jsonl_file>")
sys.exit(1)
# Initialize lists to store targets and predictions
targets = []
predictions = []
# Read the JSON lines file
filename = sys.argv[1]
with open(filename, 'r') as f:
for line in f:
obj = json.loads(line)
targets.append(obj['target'])
predictions.append(obj['prediction'])
# Joining sentences for BLEU score calculation
targets_str = ' '.join(targets)
predictions_str = ' '.join(predictions)
# Calculate BLEU score
bleu = sacrebleu.sentence_bleu(predictions_str, [targets_str])
# Print results
print(f"BLEU Score: {bleu.score}")