Bibliothèques et extensions
Découvrez des bibliothèques permettant de créer des modèles ou des méthodes avancés avec TensorFlow et accédez à des packages d'applications spécialisées pour enrichir TensorFlow de nouvelles fonctionnalités.
- Consulter la documentationExtra functionality for TensorFlow, maintained by SIG Addons.
TensorFlow Addons
- Consulter la documentationA library for designing, testing, and implementing reinforcement learning algorithms.
TensorFlow Agents
- Consulter la documentationA library to build ML models with end-to-end optimized data compression built in.
TensorFlow Compression
- Consulter la documentationA library to analyze training and serving data to compute descriptive statistics, infer schemas, and detect anomalies.
TensorFlow Data Validation
- Consulter la documentationState-of-the-art algorithms for training, serving and interpreting models that use decision forests for classification, regression and ranking.
TensorFlow Decision Forests
- A research framework for fast prototyping of reinforcement learning algorithms.
Dopamine
- Consulter la documentationA library that enables easy computation of commonly-identified fairness metrics for binary and multiclass classifiers.
Fairness Indicators
- Consulter la documentationAn open source framework for machine learning and other computations on decentralized data.
TensorFlow Federated
- A library to build neural networks on graph data (nodes and edges with arbitrary features), including tools for preparing input data and training models.
TensorFlow GNN
- Consulter la documentationA library of computer graphics functionalities ranging from cameras, lights, and materials to renderers.
TensorFlow Graphics
- Consulter la documentationA library for reusable machine learning. Download and reuse the latest trained models with a minimal amount of code.
TensorFlow Hub
- Consulter la documentationDataset, streaming, and file system extensions, maintained by SIG IO.
TensorFlow IO
- Consulter la documentationLanguage bindings for Java and other JVM languages, such as Scala or Kotlin.
TensorFlow JVM
- A library of modular components for common computer vision tasks such as data augmentation, classification, object detection, segmentation, and more.
KerasCV
- An easily customizable natural language processing library providing modular components and state-of-the-art preset weights and architectures.
KerasNLP
- Consulter la documentationA library for flexible, controlled and interpretable ML solutions with common-sense shape constraints.
TensorFlow Lattice
- Consulter la documentationA library to run ML models on digital signal processors (DSPs), microcontrollers, and other devices with limited memory.
TensorFlow Lite Micro
- Consulter la documentationA library that simplifies model training for on-device natural language processing, vision, and audio applications.
TensorFlow Lite Model Maker
- Consulter la documentationA toolkit to customize model interface on Android, create metadata, and build inference pipelines for mobile deployment.
TensorFlow Lite Support
- Utilities for passing TensorFlow-related metadata between tools.
TensorFlow Metadata
- Consulter la documentationA library for recording and retrieving MLOps metadata associated with machine learning workflows.
ML Metadata
- Consulter la documentationA library for deep analysis of model results beyond simple training metrics, to measure edge and corner cases and bias.
TensorFlow Model Analysis
- Consulter la documentationA collection of tools to generate documents that provide context and transparency into a model's development and performance.
Model Card Toolkit
- Consulter la documentationA suite of tools for optimizing ML models for deployment and execution.
Model Optimization Toolkit
- Consulter la documentationA library to help create and train models in a way that reduces or eliminates user harm resulting from underlying performance biases.
TensorFlow Model Remediation
- Utilities for manipulating data in a n-dimensional space in Java, maintained by SIG JVM.
NdArray
- Consulter la documentationA learning framework to train neural networks by leveraging structured signals in addition to feature inputs.
Neural Structured Learning
- Consulter la documentationA Python library that includes implementations of TensorFlow optimizers for training machine learning models with differential privacy.
TensorFlow Privacy
- Consulter la documentationA library for probabilistic reasoning and statistical analysis.
TensorFlow Probability
- Consulter la documentationA quantum machine learning library for rapid prototyping of hybrid quantum-classical ML models.
TensorFlow Quantum
- Consulter la documentationA library for Learning-to-Rank (LTR) techniques on the TensorFlow platform.
TensorFlow Ranking
- Consulter la documentationA library for building recommender system models.
TensorFlow Recommenders
- A collection of community projects introducing Dynamic Embedding Technology to large-scale recommendation systems built upon TensorFlow
TensorFlow Recommenders Addons
- Consulter la documentationA flexible, high-performance serving system for machine learning models, designed for production environments
TensorFlow Serving
- A library from DeepMind for constructing neural networks.
Sonnet
- Consulter la documentationA collection of text- and NLP-related classes and ops ready to use with TensorFlow 2.
TensorFlow Text
- Consulter la documentationA library for large-scale feature engineering and eliminating training-serving skew.
TensorFlow Transform
- Consulter la documentationA hardware-accelerated library for training and deploying ML models using JavaScript or Node.js.
TensorFlow.js
- Consulter la documentationAn end-to-end platform for deploying production ML pipelines.
TFX
- Consulter la documentationA collection of community projects to build new components, examples, libraries, and tools for TFX.
TFX-Addons