DotTorch.Losses
9.0.3
See the version list below for details.
dotnet add package DotTorch.Losses --version 9.0.3
NuGet\Install-Package DotTorch.Losses -Version 9.0.3
<PackageReference Include="DotTorch.Losses" Version="9.0.3" />
<PackageVersion Include="DotTorch.Losses" Version="9.0.3" />
<PackageReference Include="DotTorch.Losses" />
paket add DotTorch.Losses --version 9.0.3
#r "nuget: DotTorch.Losses, 9.0.3"
#:package DotTorch.Losses@9.0.3
#addin nuget:?package=DotTorch.Losses&version=9.0.3
#tool nuget:?package=DotTorch.Losses&version=9.0.3
DotTorch.Losses — Библиотека функций потерь для глубокого обучения на .NET
Содержание / Contents / Inhalt / 目录
Русский
DotTorch.Losses — специализированная библиотека для вычисления функций потерь в задачах машинного обучения и глубокого обучения на платформе .NET. Пакет расширяет возможности DotTorch.Core, предоставляя набор распространённых и важных loss-функций с полной поддержкой broadcasting, редукций и автоматического дифференцирования.
Основные возможности:
- Поддержка основных функций потерь: MSE, Cross-Entropy, Binary Cross-Entropy, Huber, KL Divergence, Negative Log Likelihood, Hinge Loss.
- Полная интеграция с DotTorch.Core: тензоры, broadcasting, вычислительный граф, autograd.
- Поддержка различных режимов редукции (None, Mean, Sum).
- Высокая производительность и надёжность.
- Лёгкое расширение и интеграция в пользовательские ML/AI пайплайны.
- Полное покрытие тестами для обеспечения стабильной работы.
DotTorch.Losses предназначен для эффективного и удобного вычисления потерь при обучении нейросетей и других моделей машинного обучения с использованием экосистемы .NET.
English
DotTorch.Losses is a specialized library for loss function computation in machine learning and deep learning tasks on the .NET platform. It extends DotTorch.Core by providing a comprehensive set of essential loss functions with full support for broadcasting, reduction modes, and automatic differentiation.
Key features:
- Support for core loss functions: MSE, Cross-Entropy, Binary Cross-Entropy, Huber, KL Divergence, Negative Log Likelihood, Hinge Loss.
- Full integration with DotTorch.Core: tensors, broadcasting, computation graph, autograd.
- Support for various reduction modes (None, Mean, Sum).
- High performance and reliability.
- Easy extensibility and integration into custom ML/AI pipelines.
- Comprehensive testing coverage ensuring stable operation.
DotTorch.Losses is designed for efficient and convenient loss calculation during neural network training and other ML model workflows leveraging the .NET ecosystem.
Deutsch
DotTorch.Losses ist eine spezialisierte Bibliothek zur Berechnung von Verlustfunktionen für maschinelles Lernen und Deep Learning auf der .NET-Plattform. Sie erweitert DotTorch.Core um einen umfassenden Satz essenzieller Verlustfunktionen mit voller Unterstützung für Broadcasting, Reduktionsmodi und automatische Differenzierung.
Hauptfunktionen:
- Unterstützung der wichtigsten Verlustfunktionen: MSE, Cross-Entropy, Binary Cross-Entropy, Huber, KL-Divergenz, Negative Log Likelihood, Hinge Loss.
- Volle Integration mit DotTorch.Core: Tensoren, Broadcasting, Rechengraph, Autograd.
- Unterstützung verschiedener Reduktionsmodi (None, Mean, Sum).
- Hohe Leistung und Zuverlässigkeit.
- Einfache Erweiterbarkeit und Integration in benutzerdefinierte ML/AI-Pipelines.
- Umfassende Testabdeckung für stabile Funktion.
DotTorch.Losses ist für eine effiziente und bequeme Verlustberechnung beim Training von neuronalen Netzen und anderen ML-Modell-Workflows innerhalb des .NET-Ökosystems konzipiert.
中文
DotTorch.Losses 是一个专门用于 .NET 平台机器学习和深度学习任务中损失函数计算的库。它扩展了 DotTorch.Core,提供了一套完整的核心损失函数,支持广播、归约模式和自动微分。
主要功能:
- 支持核心损失函数:MSE、交叉熵、二元交叉熵、Huber、KL散度、负对数似然、Hinge损失。
- 与 DotTorch.Core 完全集成:张量、广播、计算图、自动微分。
- 支持多种归约模式(None, Mean, Sum)。
- 高性能和可靠性。
- 易于扩展并集成到自定义的机器学习/人工智能流水线中。
- 完善的测试覆盖,确保稳定运行。
DotTorch.Losses 专为在使用 .NET 生态系统进行神经网络训练及其他机器学习模型工作流中的高效且便捷的损失计算而设计。
Product | Versions Compatible and additional computed target framework versions. |
---|---|
.NET | net8.0 is compatible. net8.0-android was computed. net8.0-browser was computed. net8.0-ios was computed. net8.0-maccatalyst was computed. net8.0-macos was computed. net8.0-tvos was computed. net8.0-windows was computed. net9.0 is compatible. net9.0-android was computed. net9.0-browser was computed. net9.0-ios was computed. net9.0-maccatalyst was computed. net9.0-macos was computed. net9.0-tvos was computed. net9.0-windows was computed. net10.0 was computed. net10.0-android was computed. net10.0-browser was computed. net10.0-ios was computed. net10.0-maccatalyst was computed. net10.0-macos was computed. net10.0-tvos was computed. net10.0-windows was computed. |
-
net8.0
- DotTorch.Core (>= 9.0.9)
-
net9.0
- DotTorch.Core (>= 9.0.9)
NuGet packages
This package is not used by any NuGet packages.
GitHub repositories
This package is not used by any popular GitHub repositories.
Version | Downloads | Last Updated | |
---|---|---|---|
9.2.4 | 151 | 7/14/2025 | |
9.2.3 | 222 | 7/14/2025 | |
9.2.2 | 220 | 7/14/2025 | |
9.2.1 | 233 | 7/13/2025 | |
9.2.0 | 218 | 7/13/2025 | |
9.1.0 | 180 | 7/13/2025 | |
9.0.14 | 149 | 7/13/2025 | |
9.0.13 | 190 | 7/12/2025 | |
9.0.12 | 174 | 7/12/2025 | |
9.0.11 | 250 | 7/10/2025 | |
9.0.10 | 318 | 7/10/2025 | |
9.0.9 | 409 | 7/9/2025 | |
9.0.8 | 407 | 7/9/2025 | |
9.0.7 | 398 | 7/9/2025 | |
9.0.6 | 401 | 7/9/2025 | |
9.0.5 | 403 | 7/9/2025 | |
9.0.4 | 399 | 7/9/2025 | |
9.0.3 | 405 | 7/9/2025 | |
9.0.2 | 405 | 7/9/2025 | |
9.0.1 | 409 | 7/8/2025 | |
9.0.0 | 411 | 7/8/2025 |
EN:
• Updated dependency DotTorch.Core from version 9.0.8 to 9.0.9.
• DotTorch.Core 9.0.9 updates include:
– Added Tensor.Uniform(shape, min, max) for random initialization.
– Extended Tensor.FromArray to support multi-dimensional arrays ([,], [,,], etc.).
– Added support for Tensor.Transpose operation with autograd support.
– Minor cleanup and performance improvements in tensor creation logic.
RU:
• Обновлена зависимость DotTorch.Core с версии 9.0.8 до 9.0.9.
• Обновления DotTorch.Core 9.0.9 включают:
– Добавлен метод Tensor.Uniform(shape, min, max) для инициализации случайными числами.
– Метод Tensor.FromArray теперь поддерживает многомерные массивы ([,], [,,] и т.д.).
– Добавлена поддержка операции транспонирования Tensor.Transpose с автоградиентом.
– Мелкие улучшения и оптимизации в логике создания тензоров.
DE:
• Abhängigkeit DotTorch.Core von Version 9.0.8 auf 9.0.9 aktualisiert.
• DotTorch.Core 9.0.9 Updates beinhalten:
– Tensor.Uniform(shape, min, max) zur Initialisierung mit Zufallswerten hinzugefügt.
– Tensor.FromArray unterstützt jetzt mehrdimensionale Arrays ([,], [,,], usw.).
– Unterstützung für Tensor.Transpose Operation mit Autograd hinzugefügt.
– Kleinere Optimierungen in der Tensor-Erstellung vorgenommen.
CN:
• 将依赖项 DotTorch.Core 从版本 9.0.8 更新至 9.0.9。
• DotTorch.Core 9.0.9 更新内容包括:
– 添加了 Tensor.Uniform(shape, min, max) 以支持随机初始化。
– Tensor.FromArray 现在支持多维数组(如 [,], [,,] 等)。
– 添加了支持自动微分的 Tensor.Transpose 操作。
– 张量构建逻辑进行了小幅优化与清理。