https://huggingface.co/learn/nlp-course/chapter1/1?fw=pt

Introduction

Welcome to the 🤗 Course!

https://youtu.be/00GKzGyWFEs?si=_fwBMxuDBpygyJSj

This course will teach you about natural language processing (NLP) using libraries from the Hugging Face ecosystem — 🤗 Transformers, 🤗 Datasets, 🤗 Tokenizers, and 🤗 Accelerate — as well as the Hugging Face Hub. It’s completely free and without ads.

이 과정에서는 Hugging Face 생태계의 라이브러리( 🤗 Transformers, 🤗 Datasets, 🤗 Tokenizers 및 🤗 Accelerate)와 Hugging Face Hub를 사용하여 자연어 처리(NLP)에 대해 설명합니다. 완전 무료이며 광고도 없습니다.

What to expect?

Here is a brief overview of the course:

강좌에 대한 간략한 개요는 다음과 같습니다.

• Chapters 1 to 4 provide an introduction to the main concepts of the 🤗 Transformers library. By the end of this part of the course, you will be familiar with how Transformer models work and will know how to use a model from the Hugging Face Hub, fine-tune it on a dataset, and share your results on the Hub!
  
  
• 1장부터 4장까지는 🤗 Transformers 라이브러리의 주요 개념을 소개합니다. 과정의 이 부분이 끝나면 Transformer 모델의 작동 방식에 익숙해지고 Hugging Face Hub의 모델을 사용하는 방법, 데이터세트에서 이를 미세 조정하고 허브에서 결과를 공유하는 방법을 알게 됩니다!
  
  
• Chapters 5 to 8 teach the basics of 🤗 Datasets and 🤗 Tokenizers before diving into classic NLP tasks. By the end of this part, you will be able to tackle the most common NLP problems by yourself.
  
  
• 5~8장에서는 고전적인 NLP 작업을 시작하기 전에 🤗 데이터세트 및 🤗 토크나이저의 기본 사항을 가르칩니다. 이 부분이 끝나면 가장 일반적인 NLP 문제를 스스로 해결할 수 있게 됩니다.
  
  
• Chapters 9 to 12 go beyond NLP, and explore how Transformer models can be used to tackle tasks in speech processing and computer vision. Along the way, you’ll learn how to build and share demos of your models, and optimize them for production environments. By the end of this part, you will be ready to apply 🤗 Transformers to (almost) any machine learning problem!
  
  
• 9~12장에서는 NLP를 넘어 음성 처리 및 컴퓨터 비전 작업을 처리하는 데 Transformer 모델을 사용할 수 있는 방법을 살펴봅니다. 그 과정에서 모델의 데모를 구축 및 공유하고 생산 환경에 맞게 최적화하는 방법을 배우게 됩니다. 이 부분이 끝나면 🤗 Transformers를 (거의) 모든 기계 학습 문제에 적용할 수 있습니다!

This course: 이 과정

• Requires a good knowledge of Python 
• Python에 대한 충분한 지식이 필요합니다.
  
  
• Is better taken after an introductory deep learning course, such as fast.ai’s Practical Deep Learning for Coders or one of the programs developed by DeepLearning.AI  
• Does not expect prior PyTorch or TensorFlow knowledge, though some familiarity with either of those will help
  
  
• fast.ai의 Practical Deep Learning for Coders 또는 DeepLearning.AI에서 개발한 프로그램 중 하나와 같은 입문 딥 러닝 과정을 수강하는 것이 좋습니다.
  
  
• PyTorch 또는 TensorFlow에 대한 사전 지식을 기대하지 않지만 둘 중 하나에 익숙하면 도움이 됩니다.

After you’ve completed this course, we recommend checking out DeepLearning.AI’s Natural Language Processing Specialization, which covers a wide range of traditional NLP models like naive Bayes and LSTMs that are well worth knowing about!

이 과정을 마친 후에는 Naive Bayes 및 LSTM과 같이 알아 둘 가치가 있는 광범위한 기존 NLP 모델을 다루는 DeepLearning.AI의 자연어 처리 전문 분야를 확인하는 것이 좋습니다!

Who are we?

About the authors: 저자 소개

Abubakar Abid completed his PhD at Stanford in applied machine learning. During his PhD, he founded Gradio, an open-source Python library that has been used to build over 600,000 machine learning demos. Gradio was acquired by Hugging Face, which is where Abubakar now serves as a machine learning team lead.

Abubakar Abid는 스탠포드에서 응용 기계 학습 분야의 박사 학위를 취득했습니다. 박사 과정 동안 그는 600,000개 이상의 기계 학습 데모를 구축하는 데 사용된 오픈 소스 Python 라이브러리인 Gradio를 설립했습니다. Gradio는 현재 Abubakar가 기계 학습 팀 리더로 일하고 있는 Hugging Face에 인수되었습니다.

Matthew Carrigan is a Machine Learning Engineer at Hugging Face. He lives in Dublin, Ireland and previously worked as an ML engineer at Parse.ly and before that as a post-doctoral researcher at Trinity College Dublin. He does not believe we’re going to get to AGI by scaling existing architectures, but has high hopes for robot immortality regardless.

Matthew Carrigan은 Hugging Face의 머신러닝 엔지니어입니다. 그는 아일랜드 더블린에 거주하며 이전에는 Parse.ly에서 ML 엔지니어로 근무했고 그 전에는 Trinity College Dublin에서 박사후 연구원으로 근무했습니다. 그는 기존 아키텍처를 확장하는 것으로는  AGI에 도달할 것이라고 믿지 않지만 그럼에도 불구하고 로봇 불멸에 대한 높은 희망을 가지고 있습니다.

Lysandre Debut is a Machine Learning Engineer at Hugging Face and has been working on the 🤗 Transformers library since the very early development stages. His aim is to make NLP accessible for everyone by developing tools with a very simple API.

Lysandre Debut는 Hugging Face의 기계 학습 엔지니어이며 초기 개발 단계부터 🤗 Transformers 라이브러리 작업을 해왔습니다. 그의 목표는 매우 간단한 API로 도구를 개발하여 모든 사람이 NLP에 액세스할 수 있도록 하는 것입니다.

Sylvain Gugger is a Research Engineer at Hugging Face and one of the core maintainers of the 🤗 Transformers library. Previously he was a Research Scientist at fast.ai, and he co-wrote Deep Learning for Coders with fastai and PyTorch with Jeremy Howard. The main focus of his research is on making deep learning more accessible, by designing and improving techniques that allow models to train fast on limited resources.

Sylvain Gugger는 Hugging Face의 연구 엔지니어이자 🤗 Transformers 라이브러리의 핵심 관리자 중 한 명입니다. 이전에 그는 fast.ai의 연구 과학자였으며, fastai와 함께 Coders를 위한 Deep Learning, Jeremy Howard와 함께 PyTorch를 공동 집필했습니다. 그의 연구의 주요 초점은 모델이 제한된 리소스에서 빠르게 훈련할 수 있는 기술을 설계하고 개선하여 딥 러닝의 접근성을 높이는 것입니다.

Dawood Khan is a Machine Learning Engineer at Hugging Face. He’s from NYC and graduated from New York University studying Computer Science. After working as an iOS Engineer for a few years, Dawood quit to start Gradio with his fellow co-founders. Gradio was eventually acquired by Hugging Face.

Dawood Khan은 Hugging Face의 머신러닝 엔지니어입니다. 그는 뉴욕 출신이고 뉴욕 대학교에서 컴퓨터 공학을 전공했습니다. 몇 년 동안 iOS 엔지니어로 일한 후 Dawood는 동료 공동 창립자들과 함께 Gradio를 시작하기 위해 회사를 그만뒀습니다. Gradio는 결국 Hugging Face에 인수되었습니다.

Merve Noyan is a developer advocate at Hugging Face, working on developing tools and building content around them to democratize machine learning for everyone.

Merve Noyan은 Hugging Face의 개발자 옹호자로서 모든 사람을 위한 기계 학습을 민주화하기 위해 도구를 개발하고 관련 콘텐츠를 구축하는 작업을 하고 있습니다.

Lucile Saulnier is a machine learning engineer at Hugging Face, developing and supporting the use of open source tools. She is also actively involved in many research projects in the field of Natural Language Processing such as collaborative training and BigScience.

Lucile Saulnier는 Hugging Face의 머신 러닝 엔지니어로, 오픈 소스 도구 사용을 개발하고 지원합니다. 그녀는 또한 협업 훈련, BigScience 등 자연어 처리 분야의 많은 연구 프로젝트에 적극적으로 참여하고 있습니다.

Lewis Tunstall is a machine learning engineer at Hugging Face, focused on developing open-source tools and making them accessible to the wider community. He is also a co-author of the O’Reilly book Natural Language Processing with Transformers.

Lewis Tunstall은 Hugging Face의 기계 학습 엔지니어로, 오픈 소스 도구를 개발하고 이를 더 넓은 커뮤니티에 액세스할 수 있도록 하는 데 중점을 두고 있습니다. 그는 또한 O'Reilly의 책인 Transformers를 사용한 자연어 처리의 공동 저자이기도 합니다.

Leandro von Werra is a machine learning engineer in the open-source team at Hugging Face and also a co-author of the O’Reilly book Natural Language Processing with Transformers. He has several years of industry experience bringing NLP projects to production by working across the whole machine learning stack..

Leandro von Werra는 Hugging Face 오픈 소스 팀의 머신 러닝 엔지니어이자 O'Reilly의 Natural Language Process with Transformers 책의 공동 저자이기도 합니다. 그는 기계 학습 스택 전반에 걸쳐 NLP 프로젝트를 프로덕션으로 가져오는 수년간의 업계 경험을 보유하고 있습니다.

FAQ

Here are some answers to frequently asked questions:

자주 묻는 질문(FAQ)에 대한 답변은 다음과 같습니다.

• Does taking this course lead to a certification? Currently we do not have any certification for this course. However, we are working on a certification program for the Hugging Face ecosystem — stay tuned!
  
  
• 이 강좌를 수강하면 인증을 받을 수 있나요? 현재 이 과정에 대한 인증이 없습니다. 그러나 우리는 Hugging Face 생태계에 대한 인증 프로그램을 개발 중입니다. 계속 지켜봐 주시기 바랍니다!
  
  
• How much time should I spend on this course? Each chapter in this course is designed to be completed in 1 week, with approximately 6-8 hours of work per week. However, you can take as much time as you need to complete the course.
  
  
• 이 강좌에 얼마나 많은 시간을 투자해야 합니까? 이 과정의 각 장은 주당 약 6~8시간씩 1주일 내에 완료하도록 설계되었습니다. 그러나 과정을 완료하는 데 필요한 만큼의 시간을 투자할 수 있습니다.
  
  
• Where can I ask a question if I have one? If you have a question about any section of the course, just click on the ”Ask a question” banner at the top of the page to be automatically redirected to the right section of the Hugging Face forums:
  
  
• 질문이 있으면 어디로 문의해야 하나요? 코스의 특정 섹션에 대해 질문이 있는 경우 페이지 상단의 "질문하기" 배너를 클릭하면 자동으로 Hugging Face 포럼의 해당 섹션으로 리디렉션됩니다.

https://discuss.huggingface.co/t/chapter-1-questions

Note that a list of project ideas is also available on the forums if you wish to practice more once you have completed the course.

과정을 마친 후 더 연습하고 싶다면 포럼에서 프로젝트 아이디어 목록을 확인할 수도 있습니다.

• Where can I get the code for the course? For each section, click on the banner at the top of the page to run the code in either Google Colab or Amazon SageMaker Studio Lab:
  
  
• 강의 코드는 어디서 받을 수 있나요? 각 섹션에 대해 페이지 상단의 배너를 클릭하여 Google Colab 또는 Amazon SageMaker Studio Lab에서 코드를 실행하세요.

The Jupyter notebooks containing all the code from the course are hosted on the huggingface/notebooks repo. If you wish to generate them locally, check out the instructions in the course repo on GitHub.

강좌의 모든 코드가 포함된 Jupyter Notebook은 Huggingface/Notebooks 저장소에서 호스팅됩니다. 로컬에서 생성하려면 GitHub의 코스 저장소에 있는 지침을 확인하세요.

• How can I contribute to the course? There are many ways to contribute to the course! If you find a typo or a bug, please open an issue on the course repo. If you would like to help translate the course into your native language, check out the instructions here.
  
  
• 강좌에 어떻게 기여할 수 있나요? 강좌에 참여하는 방법에는 여러 가지가 있습니다! 오타나 버그를 발견한 경우 코스 저장소에서 문제를 열어주세요. 강좌를 모국어로 번역하는 데 도움을 주고 싶다면 여기에서 지침을 확인하세요.
  
  
• What were the choices made for each translation? Each translation has a glossary and TRANSLATING.txt file that details the choices that were made for machine learning jargon etc. You can find an example for German here.
  
  
• 각 번역에 대해 어떤 선택이 이루어졌나요? 각 번역에는 기계 학습 전문 용어 등에 대한 선택 사항을 자세히 설명하는 용어집과 TRANSLATING.txt 파일이 있습니다. 여기에서 독일어에 대한 예를 찾을 수 있습니다.

• Can I reuse this course? Of course! The course is released under the permissive Apache 2 license. This means that you must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. If you would like to cite the course, please use the following BibTeX:
  
  
• 이 강좌를 재사용할 수 있나요? 물론! 이 과정은 허용되는 Apache 2 라이센스에 따라 공개됩니다. 이는 적절한 출처를 표시하고 라이선스에 대한 링크를 제공하고 변경 사항이 있는지 표시해야 함을 의미합니다. 귀하는 합리적인 방식으로 그렇게 할 수 있지만, 라이센스 제공자가 귀하 또는 귀하의 사용을 보증하는 방식으로 그렇게 할 수는 없습니다. 강좌를 인용하려면 다음 BibTeX를 사용하세요.

@misc{huggingfacecourse,
  author = {Hugging Face},
  title = {The Hugging Face Course, 2022},
  howpublished = "\url{https://huggingface.co/course}",
  year = {2022},
  note = "[Online; accessed <today>]"
}

Let's Go

Are you ready to roll? In this chapter, you will learn:

굴릴 준비가 되셨나요? 이 장에서는 다음 내용을 학습합니다.

• How to use the pipeline() function to solve NLP tasks such as text generation and classification
  
  
• 파이프라인() 함수를 사용하여 텍스트 생성 및 분류와 같은 NLP 작업을 해결하는 방법
  
  
• About the Transformer architecture
  
  
• Transformer 아키텍처 정보
  
  
• How to distinguish between encoder, decoder, and encoder-decoder architectures and use cases
  
  
• 인코더, 디코더, 인코더-디코더 아키텍처와 사용 사례를 구별하는 방법

https://huggingface.co/learn/nlp-course/chapter0/1?fw=pt

Introduction

Welcome to the Hugging Face course! This introduction will guide you through setting up a working environment. If you’re just starting the course, we recommend you first take a look at Chapter 1, then come back and set up your environment so you can try the code yourself.

Hugging Face 코스에 오신 것을 환영합니다! 이 소개에서는 작업 환경을 설정하는 과정을 안내합니다. 과정을 막 시작하는 경우 먼저 1장을 살펴보고 다시 돌아와서 코드를 직접 사용해 볼 수 있도록 환경을 설정하는 것이 좋습니다.

All the libraries that we’ll be using in this course are available as Python packages, so here we’ll show you how to set up a Python environment and install the specific libraries you’ll need.

이 과정에서 사용할 모든 라이브러리는 Python 패키지로 제공되므로 여기서는 Python 환경을 설정하고 필요한 특정 라이브러리를 설치하는 방법을 보여 드리겠습니다.

We’ll cover two ways of setting up your working environment, using a Colab notebook or a Python virtual environment. Feel free to choose the one that resonates with you the most. For beginners, we strongly recommend that you get started by using a Colab notebook.

Colab 노트북이나 Python 가상 환경을 사용하여 작업 환경을 설정하는 두 가지 방법을 다루겠습니다. 당신에게 가장 공감되는 것을 자유롭게 선택하십시오. 초보자의 경우 Colab 노트북을 사용하여 시작하는 것이 좋습니다.

Note that we will not be covering the Windows system. If you’re running on Windows, we recommend following along using a Colab notebook. If you’re using a Linux distribution or macOS, you can use either approach described here.

Windows 시스템은 다루지 않습니다. Windows에서 실행하는 경우 Colab 노트북을 사용하여 따라하는 것이 좋습니다. Linux 배포판이나 macOS를 사용하는 경우 여기에 설명된 접근 방식 중 하나를 사용할 수 있습니다.

Most of the course relies on you having a Hugging Face account. We recommend creating one now: create an account.

대부분의 과정은 Hugging Face 계정이 있어야 합니다. 지금 계정을 만드는 것이 좋습니다. 계정을 만드세요.

Using a Google Colab notebook

Using a Colab notebook is the simplest possible setup; boot up a notebook in your browser and get straight to coding!

Colab 노트북을 사용하는 것이 가장 간단한 설정입니다. 브라우저에서 노트북을 부팅하고 바로 코딩을 시작해 보세요!

If you’re not familiar with Colab, we recommend you start by following the introduction. Colab allows you to use some accelerating hardware, like GPUs or TPUs, and it is free for smaller workloads.

Colab에 익숙하지 않다면 소개부터 시작하는 것이 좋습니다. Colab을 사용하면 GPU 또는 TPU와 같은 일부 가속 하드웨어를 사용할 수 있으며 소규모 워크로드에는 무료로 제공됩니다.

Once you’re comfortable moving around in Colab, create a new notebook and get started with the setup:

Colab에서 편안하게 이동하고 나면 새 노트북을 만들고 설정을 시작하세요.

The next step is to install the libraries that we’ll be using in this course. We’ll use pip for the installation, which is the package manager for Python. In notebooks, you can run system commands by preceding them with the ! character, so you can install the 🤗 Transformers library as follows:

다음 단계는 이 과정에서 사용할 라이브러리를 설치하는 것입니다. Python의 패키지 관리자인 pip를 설치에 사용하겠습니다. 노트북에서는 앞에 !를 붙여 시스템 명령을 실행할 수 있습니다. 캐릭터이므로 다음과 같이 🤗 Transformers 라이브러리를 설치할 수 있습니다.

!pip install transformers

You can make sure the package was correctly installed by importing it within your Python runtime:

Python 런타임 내에서 패키지를 가져와 패키지가 올바르게 설치되었는지 확인할 수 있습니다.

import transformers

This installs a very light version of 🤗 Transformers. In particular, no specific machine learning frameworks (like PyTorch or TensorFlow) are installed. Since we’ll be using a lot of different features of the library, we recommend installing the development version, which comes with all the required dependencies for pretty much any imaginable use case:

이것은 🤗 Transformers의 매우 가벼운 버전을 설치합니다. 특히, 특정 기계 학습 프레임워크(예: PyTorch 또는 TensorFlow)가 설치되지 않습니다. 우리는 라이브러리의 다양한 기능을 사용할 것이므로 상상할 수 있는 거의 모든 사용 사례에 필요한 모든 종속성이 포함된 개발 버전을 설치하는 것이 좋습니다.

!pip install transformers[sentencepiece]

This will take a bit of time, but then you’ll be ready to go for the rest of the course!

약간의 시간이 걸리겠지만, 그러면 나머지 과정을 진행할 준비가 된 것입니다!

Using a Python virtual environment

If you prefer to use a Python virtual environment, the first step is to install Python on your system. We recommend following this guide to get started.

Python 가상 환경을 사용하려는 경우 첫 번째 단계는 시스템에 Python을 설치하는 것입니다. 시작하려면 이 가이드를 따르는 것이 좋습니다.

Once you have Python installed, you should be able to run Python commands in your terminal. You can start by running the following command to ensure that it is correctly installed before proceeding to the next steps: python --version. This should print out the Python version now available on your system.

Python을 설치하고 나면 터미널에서 Python 명령을 실행할 수 있습니다. 다음 단계를 진행하기 전에 python --version 명령을 실행하여 올바르게 설치되었는지 확인할 수 있습니다. 그러면 현재 시스템에서 사용할 수 있는 Python 버전이 인쇄됩니다.

When running a Python command in your terminal, such as python --version, you should think of the program running your command as the “main” Python on your system. We recommend keeping this main installation free of any packages, and using it to create separate environments for each application you work on — this way, each application can have its own dependencies and packages, and you won’t need to worry about potential compatibility issues with other applications.

python --version과 같은 Python 명령을 터미널에서 실행할 때 명령을 실행하는 프로그램을 시스템의 "기본" Python으로 생각해야 합니다. 이 기본 설치에 패키지를 사용하지 않고 이를 사용하여 작업하는 각 애플리케이션에 대해 별도의 환경을 만드는 것이 좋습니다. 이렇게 하면 각 애플리케이션이 자체 종속성과 패키지를 가질 수 있습니다.다른 응용 프로그램과의 잠재적인 호환성 문제에 대해 걱정할 필요가 없습니다.

In Python this is done with virtual environments, which are self-contained directory trees that each contain a Python installation with a particular Python version alongside all the packages the application needs. Creating such a virtual environment can be done with a number of different tools, but we’ll use the official Python package for that purpose, which is called venv.

Python에서는 애플리케이션에 필요한 모든 패키지와 함께 특정 Python 버전이 포함된 Python 설치가 포함된 독립형 디렉터리 트리인 가상 환경을 통해 이 작업이 수행됩니다. 이러한 가상 환경을 만드는 것은 다양한 도구를 사용하여 수행할 수 있지만 우리는 해당 목적을 위해 venv라고 하는 공식 Python 패키지를 사용하겠습니다.

First, create the directory you’d like your application to live in — for example, you might want to make a new directory called transformers-course at the root of your home directory:

먼저, 애플리케이션을 보관할 디렉터리를 만듭니다. 예를 들어 홈 디렉터리의 루트에 Transformers-course라는 새 디렉터리를 만들 수 있습니다.

mkdir ~/transformers-course
cd ~/transformers-course

From inside this directory, create a virtual environment using the Python venv module:

이 디렉터리 내에서 Python venv 모듈을 사용하여 가상 환경을 만듭니다.

python -m venv .env

나는 윈도우 환경에서 실행하고 있으며 가상 환경 이름은 hfnlp라고 하겠음.

You should now have a directory called .env in your otherwise empty folder:

이제 빈 폴더에 .env라는 디렉터리가 있어야 합니다.

ls -a

.      ..    .env

내 윈도우 로컬 환경에서는 아래와 같이 dir로 디렉토리 내용을 볼 수 있음. hfnlp 폴더가 생성 돼 있음.

You can jump in and out of your virtual environment with the activate and deactivate scripts:

활성화 및 비활성화 스크립트를 사용하여 가상 환경에 들어가고 나올 수 있습니다.

# Activate the virtual environment
source .env/bin/activate

# Deactivate the virtual environment
source .env/bin/deactivate

윈도우에서는 Scripts라는 폴더에 있는 activate를 실행하면 됨

이 가상환경에서 나오려면 deactivate 하면 됨.

You can make sure that the environment is activated by running the which python command: if it points to the virtual environment, then you have successfully activated it!

which python 명령을 실행하여 환경이 활성화되었는지 확인할 수 있습니다. 가상 환경을 가리키면 성공적으로 활성화된 것입니다!

which python

/home/<user>/transformers-course/.env/bin/python

윈도우 환경에서는 비슷한 명령어로 where python이 있음.

Installing dependencies

As in the previous section on using Google Colab instances, you’ll now need to install the packages required to continue. Again, you can install the development version of 🤗 Transformers using the pip package manager:

Google Colab 인스턴스 사용에 대한 이전 섹션과 마찬가지로 이제 계속하려면 필요한 패키지를 설치해야 합니다. 이번에도 pip 패키지 관리자를 사용하여 🤗 Transformers의 개발 버전을 설치할 수 있습니다.

pip install "transformers[sentencepiece]"

You’re now all set up and ready to go!

이제 모든 설정이 완료되었으며 사용할 준비가 되었습니다!