Tutorial Lengkap Hugging Face Transformers: Pretrained Models untuk NLP dan Vision

Hugging Face Transformers adalah library paling populer untuk state-of-the-art pretrained models. Dengan ribuan model tersedia, Anda dapat dengan mudah menggunakan BERT, GPT, T5, dan lainnya untuk klasifikasi teks, generasi, terjemahan, dan tugas computer vision.

Mengapa Hugging Face Transformers?

• Ribuan model: Akses ke 200,000+ pretrained models
• Multiple frameworks: Support PyTorch, TensorFlow, JAX
• Mudah digunakan: Simple pipelines API
• Production ready: Optimized inference
• Community aktif: Update regular dan model baru

• Text classification dan NER
• Question answering
• Text generation (GPT, LLaMA)
• Translation
• Image classification dan object detection

Instalasi

pip install transformers

Dengan PyTorch
pip install transformers torch

Dengan TensorFlow
pip install transformers tensorflow

Dengan semua extras
pip install transformers[torch,sentencepiece,tokenizers]

Verify
python -c "import transformers; print(transformers.version)"

Quick Start dengan Pipelines

1. Text Classification

from transformers import pipeline

Sentiment analysis
classifier = pipeline("sentiment-analysis")
result = classifier("Saya suka produk ini! Luar biasa!")
print(result)  # [{'label': 'POSITIVE', 'score': 0.9998}]

Multiple texts
results = classifier([
    "Ini bagus sekali!",
    "Ini sangat buruk!",
    "Biasa saja, tidak istimewa."
])

Custom model
classifier = pipeline(
    "sentiment-analysis",
    model="nlptown/bert-base-multilingual-uncased-sentiment"
)

2. Named Entity Recognition

from transformers import pipeline

ner = pipeline("ner", groupedentities=True)

result = ner("Nama saya Budi dan saya bekerja di Google di Jakarta.")
print(result)
[{'entitygroup': 'PER', 'word': 'Budi', ...},
{'entitygroup': 'ORG', 'word': 'Google', ...},

{'entitygroup': 'LOC', 'word': 'Jakarta', ...}]

3. Question Answering

from transformers import pipeline

qa = pipeline("question-answering")
context = """
Hugging Face adalah perusahaan yang mengembangkan tools untuk machine learning.
Mereka terkenal dengan library Transformers dan Hugging Face Hub.
Perusahaan ini didirikan pada tahun 2016.
"""

result = qa(
    question="Kapan Hugging Face didirikan?",
    context=context
)
print(result)  # {'answer': '2016', 'score': 0.98, ...}

4. Text Generation

from transformers import pipeline

GPT-2
generator = pipeline("text-generation", model="gpt2")
result = generator(
    "Masa depan AI adalah",
    maxlength=50,

    numreturnsequences=3

)

Dengan model lebih baik
generator = pipeline("text-generation", model="meta-llama/Llama-2-7b-hf")
result = generator(
    "Jelaskan machine learning:",
    maxnewtokens=100,

    temperature=0.7
)

from transformers import pipeline

English ke French
translator = pipeline("translationentofr")
result = translator("Hello, how are you?")
print(result)  # [{'translationtext': 'Bonjour, comment allez-vous?'}]


Multi-language
translator = pipeline("translation", model="Helsinki-NLP/opus-mt-en-id")
result = translator("Machine learning is fascinating.")

from transformers import pipeline

summarizer = pipeline("summarization")
article = """
Teks artikel panjang disini...
"""

summary = summarizer(
    article,
    maxlength=100,
    minlength=30,

    dosample=False
)
print(summary[0]['summarytext'])