8 repositorios
Implementations of error metrics used to calculate the difference between neural network predictions and targets.
Distinct from Cross-Entropy Loss Functions: Covers a general set of L1, L2, and cross-entropy losses rather than a specific domain like object detection.
Explore 8 awesome GitHub repositories matching artificial intelligence & ml · Loss Function Implementations. Refine with filters or upvote what's useful.
mmagic is a multimodal training pipeline and framework for generative AI, focusing on visual synthesis and restoration. It provides the infrastructure to build and train models for tasks such as text-to-image and text-to-video generation, 3D-aware content synthesis, and high-fidelity image translation using diffusion models and generative adversarial networks. The project distinguishes itself through specialized capabilities for generative model personalization, including techniques for fine-tuning subjects and styles. It also supports advanced visual manipulations such as latent space interp
Defines new loss modules by wrapping functional implementations in classes and registering them via configuration.
PyTorch Metric Learning is an open-source library for training neural networks to produce similarity-preserving embedding spaces. It provides a modular framework where interchangeable loss functions, mining strategies, and evaluation tools can be composed to learn representations that map similar items to nearby points and dissimilar items to distant points in the embedding space. The library distinguishes itself through a highly configurable architecture that separates concerns across several interchangeable components. Users can assemble custom loss functions from pluggable distance metrics
Trains embeddings by minimizing distance between similar pairs and maximizing distance between dissimilar pairs.
Este proyecto es un framework de aprendizaje contrastivo auto-supervisado diseñado para entrenar modelos de aprendizaje profundo para aprender representaciones visuales a partir de imágenes sin utilizar etiquetas proporcionadas por humanos. Proporciona un sistema para desarrollar modelos de representación visual preentrenados que pueden adaptarse para tareas de visión artificial posteriores. El framework incluye herramientas para la clasificación de imágenes semi-supervisada, que combina grandes conjuntos de datos sin etiquetar con pequeños conjuntos etiquetados para mejorar la precisión. También cuenta con una herramienta de evaluación de sonda lineal (linear probe) para evaluar la calidad de las características de imagen aprendidas entrenando un clasificador lineal simple sobre representaciones congeladas. El código base cubre el entrenamiento de aprendizaje profundo distribuido y la aceleración por hardware para manejar grandes tamaños de lote, junto con primitivas de optimización como la programación de tasa de aprendizaje de decaimiento de coseno y regularización de decaimiento de peso. También proporciona utilidades para la gestión de modelos, incluyendo la conversión de checkpoints preentrenados entre diferentes formatos de frameworks de aprendizaje profundo y herramientas para el despliegue de modelos. La implementación se proporciona como una colección de Jupyter Notebooks.
Analyzes the behavior of contrastive loss functions to understand their influence on visual representation learning.
This project is a high-performance C++ and CUDA neural network library designed for fast training and inference of small networks on NVIDIA GPUs. It serves as a specialized backend for neural radiance fields and coordinate-based networks, providing a fused GPU kernel library and a hash grid encoder for transforming raw input dimensions into high-dimensional representations. The library distinguishes itself through the use of C++ template metaprogramming and fused-kernel execution, which merge neural network layers into single GPU device functions to eliminate memory bottlenecks. It leverages
Computes the standard L2 loss between network predictions and targets.
This project is a collection of educational resources and reference implementations for neural network development using TensorFlow. It serves as a comprehensive learning course, machine learning curriculum, and practical implementation guide for building deep learning architectures. The codebase provides instructional materials and examples covering a wide range of model types, including convolutional neural networks for image classification, recurrent networks and long short-term memory cells for sequential data, and autoencoders for generative modeling. It also includes implementations for
Implements various loss functions to calculate the error between predictions and actual values for minimization via gradient descent.
Este proyecto es un recurso educativo integral y un manual de tutoriales para construir, entrenar y desplegar modelos de machine learning usando TensorFlow 2. Sirve como una guía de aprendizaje estructurada que cubre conceptos fundamentales de deep learning, incluyendo arquitecturas de redes neuronales, diferenciación automática y operaciones con tensores. El manual proporciona orientación técnica sobre cómo optimizar la eficiencia de ejecución mediante la gestión de memoria de GPU, entrenamiento distribuido y cuantización de modelos. También incluye guías detalladas para construir pipelines de datos de alto rendimiento y exportar modelos para servidores de producción, dispositivos móviles y navegadores web. El material abarca una amplia gama de capacidades, incluyendo el desarrollo de modelos con redes convolucionales y recurrentes, la implementación de funciones de pérdida y capas personalizadas, y el uso de modelos preentrenados para transfer learning. También aborda estrategias de despliegue para dispositivos edge y el uso de entornos de ejecución en la nube para aceleración por hardware. El recurso está implementado como una colección de Jupyter Notebooks.
Demonstrates how to implement custom mathematical loss functions to calculate error metrics during training.
This project is a machine learning educational archive and technical documentation collection. It serves as a deep learning tutorial series and implementation guide, providing theoretical explanations and practical walkthroughs for constructing and optimizing neural networks. The content focuses on the design and construction of diverse model architectures, including convolutional neural networks, Long Short-Term Memory networks, and generative adversarial networks. It details specific implementation patterns for autoencoders, sentiment analysis models, and various classification approaches.
Provides implementations of error metrics, including binary crossentropy, to quantify the difference between predictions and targets.
Lightly is a self-supervised learning framework and computer vision data curation tool designed to manage large image datasets and train models on unlabeled data. It functions as a PyTorch vision library and dataset management SDK, providing tools to convert raw images into high-dimensional vectors for similarity search, visualization, and feature extraction. The project implements a variety of self-supervised architectures, including MoCo, SimCLR, VICReg, Barlow Twins, and masked image modeling. It distinguishes itself by combining these learning frameworks with active learning capabilities,
Implements specialized contrastive loss functions and memory banks to store past examples as negative samples.