The code for table 1.

[ShiYaya]

Thank you very much for releasing the code, but I found that it is not complete, can you provide the code for table 1. Regarding the correlation coefficients of those automatic evaluation metrics, I can't get the same results as you. It may be that I have an error somewhere. So I hope you can provide your code.

[ShiYaya]

When you compute Kendall’s tau (τ) and Spearman’s rho (ρ) rank correlation, whether you retain decimal places? Because I found that keep different decimal places and get different results. But all are different from you!

Keep 5 decimal places： spearman:{'Bleu_1': 0.4482, 'Bleu_4': 0.3238, 'METEOR': 0.5866, 'ROUGE_L': 0.4468, 'CIDEr': 0.6061} kendalltau:{'Bleu_1': 0.3409, 'Bleu_4': 0.2854, 'METEOR': 0.4497, 'ROUGE_L': 0.3359, 'CIDEr': 0.4699}

Keep 4 decimal places： spearman:{'Bleu_1': 0.4482, 'Bleu_4': 0.2458, 'METEOR': 0.5866, 'ROUGE_L': 0.4468, 'CIDEr': 0.6065} kendalltau:{'Bleu_1': 0.3409, 'Bleu_4': 0.2183, 'METEOR': 0.4497, 'ROUGE_L': 0.3359, 'CIDEr': 0.4716}

Keep 3 decimal places： spearman:{'Bleu_1': 0.4482, 'Bleu_4': 0.1608, 'METEOR': 0.5865, 'ROUGE_L': 0.4468, 'CIDEr': 0.6067} kendalltau:{'Bleu_1': 0.341, 'Bleu_4': 0.1436, 'METEOR': 0.4507, 'ROUGE_L': 0.3361, 'CIDEr': 0.4767}