探索向量空間

Introduction to Embeddings with the OpenAI API

Emmanuel Pire

Senior Software Engineer, DataCamp

範例：為標題建立嵌入向量

articles = [
    {"headline": "Economic Growth Continues Amid Global Uncertainty", "topic": "Business"},
    {"headline": "Interest rates fall to historic lows", "topic": "Business"},
    {"headline": "Scientists Make Breakthrough Discovery in Renewable Energy", "topic": "Science"},
    {"headline": "India Successfully Lands Near Moon's South Pole", "topic": "Science"},
    {"headline": "New Particle Discovered at CERN", "topic": "Science"},
    {"headline": "Tech Company Launches Innovative Product to Improve Online Accessibility", "topic": "Tech"},
    {"headline": "Tech Giant Buys 49% Stake In AI Startup", "topic": "Tech"},
    {"headline": "New Social Media Platform Has Everyone Talking!", "topic": "Tech"},
    {"headline": "The Blues get promoted on the final day of the season!", "topic": "Sport"},
    {"headline": "1.5 Billion Tune-in to the World Cup Final", "topic": "Sport"}
]

範例：為標題建立嵌入向量

包含標題及其嵌入向量的字典清單。

同時為多筆輸入建立嵌入向量

headline_text = [article['headline'] for article in articles]
headline_text

["Economic Growth Continues Amid Global Uncertainty",
 ...,
 "1.5 Billion Tune-in to the World Cup Final"]

response = client.embeddings.create(
  model="text-embedding-3-small",
  input=headline_text
)
response_dict = response.model_dump()

批次處理 比多次呼叫 API 更有效率

[...]

'data': [
    {
      "embedding": [-0.017142612487077713, ..., -0.0012911480152979493],
      "index": 0,
      "object": "embedding"
    },
    {
      "embedding": [-0.032995883375406265, ..., -0.0028605300467461348],
      "index": 1,
      "object": "embedding"
    },
    ...
  ]

[...]

同時為多筆輸入建立嵌入向量

articles = [
    {"headline": "Economic Growth Continues Amid Global Uncertainty", "topic": "Business"},
     ...
]

for i, article in enumerate(articles):

    article['embedding'] = response_dict['data'][i]['embedding']


print(articles[:2])

[{'headline': 'Economic Growth Continues Amid Global Uncertainty',
  'topic': 'Business',
  'embedding': [-0.017142612487077713, ..., -0.0012911480152979493]}
 {'headline': 'Interest rates fall to historic lows',
  'topic': 'Business',
  'embedding': [-0.032995883375406265, ..., -0.0028605300467461348]}]

嵌入向量的長度是多少？

「Economic Growth Continues Amid Global Uncertainty」

len(articles[0]['embedding'])

「Tech Company Launches Innovative Product to Improve Accessibility」

len(articles[5]['embedding'])

永遠回傳 1536 個數值！

降維與 t-SNE

有多種技術可用來「降維」
t-SNE（t-distributed Stochastic Neighbor Embedding）

¹ https://www.datacamp.com/tutorial/introduction-t-sne

實作 t-SNE

from sklearn.manifold import TSNE
import numpy as np


embeddings = [article['embedding'] for article in articles]


tsne = TSNE(n_components=2, perplexity=5)

embeddings_2d = tsne.fit_transform(np.array(embeddings))

n_components：結果維度數
perplexity：演算法用的參數，必須小於資料點數
會造成部分資訊流失

¹ https://www.datacamp.com/tutorial/introduction-t-sne

視覺化嵌入向量

import matplotlib.pyplot as plt

plt.scatter(embeddings_2d[:, 0], embeddings_2d[:, 1])


topics = [article['topic'] for article in articles]
for i, topic in enumerate(topics):
    plt.annotate(topic, (embeddings_2d[i, 0], embeddings_2d[i, 1]))

plt.show()

視覺化嵌入向量

類似的文章會被聚在一起！
模型捕捉到語意

接下來：計算相似度

2D 向量空間圖，顯示相同情感與主題的評論在向量空間中更靠近。

一起來練習吧！

Introduction to Embeddings with the OpenAI API