컬렉션을 사용해 정리하기
내 환경설정을 기준으로 콘텐츠를 저장하고 분류하세요.
TensorFlow 소개
TensorFlow를 사용하면 초보자와 전문가 모두 데스크톱, 모바일, 웹 및 클라우드용 머신러닝 모델을 쉽게 만들 수 있습니다. 시작하려면 아래 섹션을 참조하세요.
TensorFlow
다음 머신러닝 프로젝트를 만드는 데 도움이 되는 초보자 및 전문가용 튜토리얼로 TensorFlow의 기본 내용을 학습하세요.
웹용
TensorFlow.js를 사용하여 새로운 머신러닝 모델을 만들고 자바스크립트로 기존 모델을 배포하세요.
모바일 및 에지용
Run inference with LiteRT on mobile and embedded devices like Android, iOS, Edge TPU, and Raspberry Pi.
프로덕션용
TFX를 사용하여 프로덕션에서 바로 사용 가능한 학습 및 추론용 ML 파이프라인을 배포하세요.
성공적인 ML 결과를 얻기 위해 데이터 준비 및 로드하기
힘들게 만든 ML 모델의 성공 여부를 가르는 가장 중요한 요소는 바로 데이터입니다.
TensorFlow는 규모에 따라 데이터를 통합, 정리, 전처리할 수 있도록 여러 가지 데이터 관련 도구를 제공합니다.
또한 책임감 있는 AI 도구를 사용하면 데이터에 포함된 편향을 발견하고 제거함으로써 모델을 통해 공정하며 윤리적인 결과를 얻을 수 있습니다.
TensorFlow 생태계로 모델 빌드 및 미세 조정하기
모델 구축, 학습, 내보내기를 간소화하는 핵심 프레임워크를 토대로 구축된 전체 생태계를 살펴보세요. TensorFlow는 Keras를 사용한 분산 학습, 즉각적인 모델 반복, 간편한 디버깅 등 다양한 기능을 지원합니다. 모델 분석 및 텐서보드와 같은 도구를 사용하면 모델의 수명 주기에 따라 개발 및 개선 상황을 추적할 수 있습니다.
TensorFlow를 시작하려면 TensorFlow Hub에서 Google 및 커뮤니티에서 제공한 선행 학습된 모델 모음을 찾아보세요. 아니면 모델 가든에서 최신 연구 모델이 어떻게 구현되었는지 살펴봐도 됩니다. 이러한 고급 구성요소 라이브러리에서 강력한 성능의 모델을 활용할 수 있으며, 새로운 데이터로 모델을 미세 조정하거나 맞춤설정하여 새로운 작업을 수행하게 할 수도 있습니다.
기기, 브라우저, 온프레미스, 클라우드에 모델 배포하기
TensorFlow provides robust capabilities to deploy your models on any environment - servers, edge devices, browsers, mobile, microcontrollers, CPUs, GPUs, FPGAs. TensorFlow Serving can run ML models at production scale on the most advanced processors in the world, including Google's custom Tensor Processing Units (TPUs).
If you need to analyze data close to its source to reduce latency and improve data privacy, the LiteRT framework lets you run models on mobile devices, edge computing devices, and even microcontrollers, and the TensorFlow.js framework lets you run machine learning with just a web browser.
프로덕션 ML을 위한 MLOps 구현
TensorFlow 플랫폼에서 데이터 자동화, 모델 추적, 성능 모니터링, 모델 재학습을 위한 권장사항을 구현하세요.
성공을 거두려면 프로덕션급 도구를 사용하여 제품, 서비스, 비즈니스 프로세스의 전체 기간에 걸쳐 모델 학습을 자동화하고 추적해야 합니다. TFX는 전체 MLOps 배포를 위한 소프트웨어 프레임워크와 도구를 제공하므로 시간이 지남에 따라 진화하는 데이터와 모델에서 문제를 감지할 수 있습니다.
ML 알아보기
엄선된 커리큘럼으로 기본적인 ML 분야의 역량을 키워보세요.
[null,null,[],[],[],null,["# Introduction to TensorFlow\n==========================\n\nTensorFlow makes it easy for beginners and experts to create machine learning models for desktop, mobile, web, and cloud. See the sections below to get started. \n\n#### TensorFlow\n\nLearn the foundations of TensorFlow with tutorials for beginners and experts to help you create your next machine learning project. \n[Learn more](/tutorials) \n\n#### For Web\n\nUse TensorFlow.js to create new machine learning models and deploy existing models with JavaScript. \n[Learn more](/js) \n\n#### For Mobile \\& Edge\n\nRun inference with LiteRT on mobile and embedded devices like Android, iOS, Edge TPU, and Raspberry Pi. \n[Learn more](https://ai.google.dev/edge/litert) \n\n#### For Production\n\nDeploy a production-ready ML pipeline for training and inference using TFX. \n[Learn more](/tfx) \n\nAn end-to-end platform for machine learning\n-------------------------------------------\n\n### Prepare and load data for successful ML outcomes\n\nData can be the most important factor in the success of your ML endeavors.\nTensorFlow offers multiple data tools to help you consolidate, clean and preprocess data at scale:\n\n- [Standard datasets](https://www.tensorflow.org/datasets) for initial training and validation\n- Highly scalable [data pipelines](https://www.tensorflow.org/guide/data) for loading data\n- [Preprocessing layers](https://www.tensorflow.org/guide/keras/preprocessing_layers) for common input transformations\n- Tools to [validate](https://www.tensorflow.org/tfx/guide/tfdv) and [transform](https://www.tensorflow.org/tfx/guide/tft) large datasets\n\nAdditionally, [responsible AI](https://www.tensorflow.org/responsible_ai) tools help you uncover and eliminate bias in your data to produce fair, ethical outcomes from your models. \n\n#### Try it in Colab\n\n[Load and preprocess an image dataset](https://colab.research.google.com/github/tensorflow/docs/blob/master/site/en/tutorials/load_data/images.ipynb) [Investigate and visualize datasets](https://colab.research.google.com/github/tensorflow/tfx/blob/master/docs/tutorials/data_validation/tfdv_basic.ipynb) \n\n### Build and fine-tune models with the TensorFlow ecosystem\n\nExplore an entire ecosystem built on the [Core framework](https://www.tensorflow.org/guide/basics) that streamlines model construction, training, and export. TensorFlow supports distributed training, immediate model iteration and easy debugging with [Keras](https://keras.io/), and much more. Tools like [Model Analysis](https://www.tensorflow.org/tfx/guide/tfma) and [TensorBoard](https://www.tensorflow.org/tensorboard) help you track development and improvement through your model's lifecycle. \n\nTo help you get started, find collections of pre-trained models at [TensorFlow Hub](https://www.tensorflow.org/hub) from Google and the community, or implementations of state-of-the art research models in the [Model Garden](https://www.tensorflow.org/guide/model_garden). These libraries of high level components allow you to take powerful models, and fine-tune them on new data or customize them to perform new tasks. \n\n#### Try it in Colab\n\n[Train a neural network to classify images](https://colab.research.google.com/github/tensorflow/docs/blob/master/site/en/tutorials/keras/classification.ipynb) [Retrain an image classifier with transfer learning](https://colab.research.google.com/github/tensorflow/hub/blob/master/examples/colab/tf2_image_retraining.ipynb) \n\n### Deploy models on-device, in the browser, on-prem, or in the cloud\n\nTensorFlow provides robust capabilities to deploy your models on any environment - servers, edge devices, browsers, mobile, microcontrollers, CPUs, GPUs, FPGAs. [TensorFlow Serving](https://www.tensorflow.org/tfx/guide/serving) can run ML models at production scale on the most advanced processors in the world, including Google's custom Tensor Processing Units (TPUs). \n\nIf you need to analyze data close to its source to reduce latency and improve data privacy, the [LiteRT](https://ai.google.dev/edge/litert/guide#1_choose_a_model) framework lets you run models on mobile devices, edge computing devices, and even microcontrollers, and the [TensorFlow.js](https://www.tensorflow.org/js) framework lets you run machine learning with just a web browser. \n\n#### Try it in Colab\n\n[Serve a model with TensorFlow Serving](https://colab.research.google.com/github/tensorflow/tfx/blob/master/docs/tutorials/serving/rest_simple.ipynb) \n\n### Implement MLOps for production ML\n\nThe TensorFlow platform helps you implement best practices for data automation, model tracking, performance monitoring, and model retraining. \n\nUsing production-level tools to automate and track model training over the lifetime of a product, service, or business process is critical to success. [TFX](https://www.tensorflow.org/tfx) provides software frameworks and tooling for full MLOps deployments, detecting issues as your data and models evolve over time. \n\n#### Try it in Colab\n\n[Create and run a simple TFX pipeline](https://colab.research.google.com/github/tensorflow/tfx/blob/master/docs/tutorials/tfx/penguin_simple.ipynb) [Track lineage with ML Metadata](https://colab.research.google.com/github/tensorflow/tfx/blob/master/docs/tutorials/mlmd/mlmd_tutorial.ipynb) \n\nLooking to expand your ML knowledge?\n------------------------------------\n\nTensorFlow is easier to use with a basic understanding of machine learning principles and core concepts. Learn and apply fundamental machine learning practices to develop your skills. \n[Learn ML](/resources/learn-ml) \nBegin with curated curriculums to improve your skills in foundational ML areas. \n[Learn more](/resources/learn-ml) \n\nGet started with TensorFlow\n---------------------------\n\n[Explore tutorials](/tutorials)"]]
