calculator-main 1.0.1

Calculate the BLEU, ROUGE, PPL scores of large models

BLEU ROUGE PPL 评分计算工具包

安装

pip install calculator_main

使用

from calculator_main.main import LLMAssessment

示例

# 示例：困惑度计算
sentences = [
    ['I', 'have', 'a', 'pen'],
    ['He', 'has', 'a', 'book'],
    ['She', 'has', 'a', 'cat']
]
unigram = {
    'I': 1 / 12, 'have': 1 / 12, 'a': 3 / 12, 'pen': 1 / 12,
    'He': 1 / 12, 'has': 2 / 12, 'book': 1 / 12, 'She': 1 / 12, 'cat': 1 / 12
}
perplexity = LLMAssessment.calculate_perplexity(sentences, unigram)
print("困惑度为：", perplexity)

# 示例：ROUGE分数计算
generated_text = "This is some generated text."
reference_text = "This is a reference text."
rouge_scores = LLMAssessment.calculate_rouge(generated_text, reference_text)
print("ROUGE分数：", rouge_scores)

# 示例：BLEU分数计算
candidate_text = ["This", "is", "some", "generated", "text"]
reference_texts = [
    ["This", "is", "a", "reference", "text"],
    ["This", "is", "another", "reference", "text"]
]
bleu_scores = LLMAssessment.calculate_bleu(reference_texts, candidate_text)
print("BLEU分数：", bleu_scores)