Similarities：精准类似度核算与语义匹配查找工具包，多维度实现多种算法，掩盖文本、图画等范畴，支撑文搜、图搜文、图搜图匹配查找

Similarities 类似度核算、语义匹配查找工具包，实现了多种类似度核算、匹配查找算法，支撑文本、图画等。

1. 文本类似度核算（文本匹配）

• 余弦类似（Cosine Similarity）：两向量求余弦
• 点积（Dot Product）：两向量归一化后求内积
• 汉明间隔（Hamming Distance），编辑间隔（Levenshtein Distance），欧氏间隔（Euclidean Distance），曼哈顿间隔（Manhattan Distance）等

• 语义模型

  • CoSENT文本匹配模型【引荐】
  • BERT模型（文本向量表征）
  • SentenceBERT文本匹配模型

• 字面模型

  • Word2Vec文本浅层语义表征【引荐】
  • 近义词词林
  • 知网Hownet义原匹配
  • BM25、RankBM25
  • TFIDF
  • SimHash

2.图画类似度核算（图画匹配）

• 语义模型

  • CLIP(Contrastive Language-Image Pre-Training)
  • VGG(doing)
  • ResNet(doing)

• 特征提取

  • pHash【引荐】, dHash, wHash, aHash
  • SIFT, Scale Invariant Feature Transform(SIFT)
  • SURF, Speeded Up Robust Features(SURF)(doing)

3.图文类似度核算

• CLIP(Contrastive Language-Image Pre-Training)

4.匹配查找

• SemanticSearch：向量类似检索，运用Cosine Similarty + topk高效核算，比1对1暴力核算快一个数量级

6.Demo展现

Compute similarity score Demo: huggingface.co/spaces/shib…

Semantic Search Demo: huggingface.co/spaces/shib…

6.1 中文文本匹配模型评测成果

成果值运用spearman系数

Model:

• Cilin
• Hownet
• SimHash
• TFIDF

• Install

pip3 install torch # conda install pytorch
pip3 install -U similarities

or

git clone https://github.com/shibing624/similarities.git
cd similarities
python3 setup.py install

7.运用场景引荐

7.1. 文本语义类似度核算

example: examples/text_similarity_demo.py

from similarities import Similarity
m = Similarity()
r = m.similarity('怎么替换花呗绑定银行卡', '花呗更改绑定银行卡')
print(f"similarity score: {float(r)}")  # similarity score: 0.855146050453186

Similarity的默许办法：

Similarity(corpus: Union[List[str], Dict[str, str]] = None,
           model_name_or_path="shibing624/text2vec-base-chinese",
           max_seq_length=128)

• 返回值：余弦值score规模是[-1, 1]，值越大越类似
• corpus：查找用的doc集，仅查找时需求，输入格式：语句列表List[str]或者{corpus_id: sentence}的Dict[str, str]格式
• model_name_or_path：模型名称或者模型路径，默许会从HF model hub下载并运用中文语义匹配模型shibing624/text2vec-base-chinese，如果是多语言景，可以替换为多语言匹配模型shibing624/text2vec-base-multilingual
• max_seq_length：输入语句的最大长度，最大为匹配模型支撑的最大长度，BERT系列是512

7.2. 文本语义匹配查找

一般在文档候选集中找与query最类似的文本，常用于QA场景的问句类似匹配、文本类似检索等使命。

example: examples/text_semantic_search_demo.py

import sys
sys.path.append('..')
from similarities import Similarity
#1.Compute cosine similarity between two sentences.
sentences = ['怎么替换花呗绑定银行卡',
             '花呗更改绑定银行卡']
corpus = [
    '花呗更改绑定银行卡',
    '我什么时候开通了花呗',
    '俄罗斯正告乌克兰对立欧盟协议',
    '暴风雨掩埋了东北部；新泽西16英寸的降雪',
    '中央情报局局长拜访以色列叙利亚谈判',
    '人在巴基斯坦基地的炸弹袭击中丧生',
]
model = Similarity(model_name_or_path="shibing624/text2vec-base-chinese")
print(model)
similarity_score = model.similarity(sentences[0], sentences[1])
print(f"{sentences[0]} vs {sentences[1]}, score: {float(similarity_score):.4f}")
print('-' * 50 + '\n')
#2.Compute similarity between two list
similarity_scores = model.similarity(sentences, corpus)
print(similarity_scores.numpy())
for i in range(len(sentences)):
    for j in range(len(corpus)):
        print(f"{sentences[i]} vs {corpus[j]}, score: {similarity_scores.numpy()[i][j]:.4f}")
print('-' * 50 + '\n')
#3.Semantic Search
model.add_corpus(corpus)
res = model.most_similar(queries=sentences, topn=3)
print(res)
for q_id, c in res.items():
    print('query:', sentences[q_id])
    print("search top 3:")
    for corpus_id, s in c.items():
        print(f'\t{model.corpus[corpus_id]}: {s:.4f}')

output:

怎么替换花呗绑定银行卡 vs 花呗更改绑定银行卡, score: 0.8551
...
怎么替换花呗绑定银行卡 vs 花呗更改绑定银行卡, score: 0.8551
怎么替换花呗绑定银行卡 vs 我什么时候开通了花呗, score: 0.7212
怎么替换花呗绑定银行卡 vs 俄罗斯正告乌克兰对立欧盟协议, score: 0.1450
怎么替换花呗绑定银行卡 vs 暴风雨掩埋了东北部；新泽西16英寸的降雪, score: 0.2167
怎么替换花呗绑定银行卡 vs 中央情报局局长拜访以色列叙利亚谈判, score: 0.2517
怎么替换花呗绑定银行卡 vs 人在巴基斯坦基地的炸弹袭击中丧生, score: 0.0809
花呗更改绑定银行卡 vs 花呗更改绑定银行卡, score: 1.0000
花呗更改绑定银行卡 vs 我什么时候开通了花呗, score: 0.6807
花呗更改绑定银行卡 vs 俄罗斯正告乌克兰对立欧盟协议, score: 0.1714
花呗更改绑定银行卡 vs 暴风雨掩埋了东北部；新泽西16英寸的降雪, score: 0.2162
花呗更改绑定银行卡 vs 中央情报局局长拜访以色列叙利亚谈判, score: 0.2728
花呗更改绑定银行卡 vs 人在巴基斯坦基地的炸弹袭击中丧生, score: 0.1279
query: 怎么替换花呗绑定银行卡
search top 3:
	花呗更改绑定银行卡: 0.8551
	我什么时候开通了花呗: 0.7212
	中央情报局局长拜访以色列叙利亚谈判: 0.2517

余弦score的值规模[-1, 1]，值越大，表明该query与corpus的文本越类似。

7.2.1 多语言文本语义类似度核算和匹配查找

多语言：包含中、英、韩、日、德、意等多国语言

example: examples/text_semantic_search_multilingual_demo.py

7.3. 快速近似文本语义匹配查找

支撑Annoy、Hnswlib的近似语义匹配查找，常用于百万数据集的匹配查找使命。

example: examples/fast_text_semantic_search_demo.py

7.4. 基于字面的文本类似度核算和匹配查找

支撑近义词词林（Cilin）、知网Hownet、词向量（WordEmbedding）、Tfidf、SimHash、BM25等算法的类似度核算和字面匹配查找，常用于文本匹配冷启动。

example: examples/literal_text_semantic_search_demo.py

from similarities import SimHashSimilarity, TfidfSimilarity, BM25Similarity, \
    WordEmbeddingSimilarity, CilinSimilarity, HownetSimilarity
text1 = "怎么替换花呗绑定银行卡"
text2 = "花呗更改绑定银行卡"
corpus = [
    '花呗更改绑定银行卡',
    '我什么时候开通了花呗',
    '俄罗斯正告乌克兰对立欧盟协议',
    '暴风雨掩埋了东北部；新泽西16英寸的降雪',
    '中央情报局局长拜访以色列叙利亚谈判',
    '人在巴基斯坦基地的炸弹袭击中丧生',
]
queries = [
    '我的花呗开通了？',
    '乌克兰被俄罗斯正告'
]
m = TfidfSimilarity()
print(text1, text2, ' sim score: ', m.similarity(text1, text2))
m.add_corpus(corpus)
res = m.most_similar(queries, topn=3)
print('sim search: ', res)
for q_id, c in res.items():
    print('query:', queries[q_id])
    print("search top 3:")
    for corpus_id, s in c.items():
        print(f'\t{m.corpus[corpus_id]}: {s:.4f}')

output:

怎么替换花呗绑定银行卡 花呗更改绑定银行卡  sim score:  0.8203384355246909
sim search:  {0: {2: 0.9999999403953552, 1: 0.43930041790008545, 0: 0.0}, 1: {0: 0.7380483150482178, 1: 0.0, 2: 0.0}}
query: 我的花呗开通了？
search top 3:
	我什么时候开通了花呗: 1.0000
	花呗更改绑定银行卡: 0.4393
	俄罗斯正告乌克兰对立欧盟协议: 0.0000
...

7.5. 图画类似度核算和匹配查找

支撑CLIP、pHash、SIFT等算法的图画类似度核算和匹配查找。

example: examples/image_semantic_search_demo.py

import sys
import glob
from PIL import Image
sys.path.append('..')
from similarities import ImageHashSimilarity, SiftSimilarity, ClipSimilarity
def sim_and_search(m):
    print(m)
    # similarity
    sim_scores = m.similarity(imgs1, imgs2)
    print('sim scores: ', sim_scores)
    for (idx, i), j in zip(enumerate(image_fps1), image_fps2):
        s = sim_scores[idx] if isinstance(sim_scores, list) else sim_scores[idx][idx]
        print(f"{i} vs {j}, score: {s:.4f}")
    # search
    m.add_corpus(corpus_imgs)
    queries = imgs1
    res = m.most_similar(queries, topn=3)
    print('sim search: ', res)
    for q_id, c in res.items():
        print('query:', image_fps1[q_id])
        print("search top 3:")
        for corpus_id, s in c.items():
            print(f'\t{m.corpus[corpus_id].filename}: {s:.4f}')
    print('-' * 50 + '\n')
image_fps1 = ['data/image1.png', 'data/image3.png']
image_fps2 = ['data/image12-like-image1.png', 'data/image10.png']
imgs1 = [Image.open(i) for i in image_fps1]
imgs2 = [Image.open(i) for i in image_fps2]
corpus_fps = glob.glob('data/*.jpg') + glob.glob('data/*.png')
corpus_imgs = [Image.open(i) for i in corpus_fps]
#2.image and image similarity score
sim_and_search(ClipSimilarity())  # the best result
sim_and_search(ImageHashSimilarity(hash_function='phash'))
sim_and_search(SiftSimilarity())

output:

Similarity: ClipSimilarity, matching_model: CLIPModel
sim scores:  tensor([[0.9580, 0.8654],
        [0.6558, 0.6145]])
data/image1.png vs data/image12-like-image1.png, score: 0.9580
data/image3.png vs data/image10.png, score: 0.6145
sim search:  {0: {6: 0.9999999403953552, 0: 0.9579654932022095, 4: 0.9326782822608948}, 1: {8: 0.9999997615814209, 4: 0.6729235649108887, 0: 0.6558331847190857}}
query: data/image1.png
search top 3:
	data/image1.png: 1.0000
	data/image12-like-image1.png: 0.9580
	data/image8-like-image1.png: 0.9327

7.6. 图文互搜

CLIP 模型不只支撑以图搜图，还支撑中英文图文互搜：

import sys
import glob
from PIL import Image
sys.path.append('..')
from similarities import ImageHashSimilarity, SiftSimilarity, ClipSimilarity
m = ClipSimilarity()
print(m)
#similarity score between text and image
image_fps = ['data/image3.png',  # yellow flower image
             'data/image1.png']  # tiger image
texts = ['a yellow flower', '老虎']
imgs = [Image.open(i) for i in image_fps]
sim_scores = m.similarity(imgs, texts)
print('sim scores: ', sim_scores)
for (idx, i), j in zip(enumerate(image_fps), texts):
    s = sim_scores[idx][idx]
    print(f"{i} vs {j}, score: {s:.4f}")

output:

sim scores:  tensor([[0.3220, 0.2409],
        [0.1677, 0.2959]])
data/image3.png vs a yellow flower, score: 0.3220
data/image1.png vs 老虎, score: 0.2112

参阅链接:github.com/shibing624/…

更多优质内容请关注公号：汀丶人工智能；会提供一些相关的资源和优质文章，免费获取阅览。