Menyelidiki ruang vektor

Pengantar Embeddings dengan OpenAI API

Emmanuel Pire

Senior Software Engineer, DataCamp

Contoh: Embedding judul berita

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"}
]

Contoh: Embedding judul berita

Daftar kamus yang berisi judul dan embedding-nya.

Embedding banyak input

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()

Batching lebih efisien daripada banyak panggilan API

[...]

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

[...]

Embedding banyak input

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]}]

Seberapa panjang vektor embedding?

"Pertumbuhan Ekonomi Berlanjut di Tengah Ketidakpastian Global"

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

"Perusahaan Teknologi Luncurkan Produk Inovatif untuk Meningkatkan Aksesibilitas"

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

Selalu mengembalikan 1536 angka!

Reduksi dimensi dan t-SNE

Beragam teknik untuk mengurangi jumlah dimensi
t-SNE (t-distributed Stochastic Neighbor Embedding)

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

Mengimplementasikan 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: jumlah dimensi hasil
perplexity: parameter algoritme, harus kurang dari jumlah data
Akan menyebabkan kehilangan informasi

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

Memvisualisasikan embedding

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()

Memvisualisasikan embedding

Artikel yang mirip dikelompokkan bersama!
Model menangkap makna semantik

Berikutnya: Menghitung kemiripan

Plot ruang vektor 2D yang menunjukkan ulasan dengan sentimen dan topik yang sama dipetakan lebih berdekatan di ruang vektor.

Ayo berlatih!

Pengantar Embeddings dengan OpenAI API