如何实现TF-IDF评分，并对某些术语进行额外加权

import math
from collections import Counter

def compute_tf(word_counts):
    total_words = sum(word_counts.values())
    return {word: count / total_words for word, count in word_counts.items()}

def compute_idf(documents):
    doc_count = len(documents)
    word_doc_counts = Counter()
    for doc in documents:
        word_doc_counts.update(set(doc))
    
    idf = {word: math.log((doc_count + 1) / (count + 1)) for word, count in word_doc_counts.items()}
    return idf

def compute_tfidf(documents, extra_weights=None):
    tfidf_scores = []
    idf = compute_idf(documents)
    
    for doc in documents:
        word_counts = Counter(doc)
        tf = compute_tf(word_counts)
        
        doc_tfidf = {}
        for word, count in word_counts.items():
            weight = extra_weights.get(word, 1) if extra_weights else 1
            doc_tfidf[word] = tf[word] * idf[word] * weight
        
        tfidf_scores.append(doc_t’tfidf)
    
    return tfidf_scores

# 示例文档
documents = [
    ["apple", "banana", "apple", "orange"],
    ["banana", "orange", "banana"],
    ["apple", "apple", "grape"]
]

# 额外加权术语
extra_weights = {
    "apple": 2,
    "grape": 1.5
}

tfidf_scores = compute_tfidf(documents, extra_weights)
for i, scores in enumerate(tfidf_scores):
    print(f"Document {i + 1} TF-IDF scores:")
    for word, score in scores.items():
        print(f"  {word}: {score:.4f}")