14 repositorios
Models capable of categorizing data into classes not seen during the training phase using semantic descriptions.
Distinguishing note: Focuses on inference-time classification without task-specific fine-tuning.
Explore 14 awesome GitHub repositories matching artificial intelligence & ml · Zero-Shot Classification Models. Refine with filters or upvote what's useful.
CLIP is a neural network architecture designed to map visual and textual data into a shared latent vector space. By utilizing transformer-based feature extraction and multi-modal tokenization, the system aligns images and natural language strings, enabling cross-modal similarity analysis and semantic classification. The project functions as a zero-shot classification engine, identifying image content by calculating the cosine similarity between visual features and arbitrary text labels without requiring task-specific retraining. Beyond inference, it serves as a research toolkit for evaluating
Identifying the content of images by comparing them against arbitrary text descriptions without needing to train custom models for specific categories.
Open CLIP is an open source framework for training and deploying Contrastive Language-Image Pre-training models. It serves as a vision-language training framework and multimodal embedding engine that maps images and text into a shared vector space for similarity searches and zero-shot classification. The project provides a toolkit for distributed training of contrastive models and includes an image-to-text generative model for producing natural language descriptions. It supports custom text encoder integration and utilizes teacher-student model distillation to transfer knowledge from large pr
Enables categorization of images using text prompts without task-specific label training.
This project is a research library and toolkit for deep learning computer vision, focused on implementing transformer and mixer-based architectures for image classification. It processes visual data by converting images into sequences of patches, allowing standard attention mechanisms to capture global dependencies without relying on traditional convolutional operations. The framework distinguishes itself through its support for multimodal embedding analysis, which maps images and text into a shared latent vector space. This capability enables zero-shot classification and cross-modal retrieva
Enables zero-shot classification by calculating similarity between image and text embeddings without task-specific training.
This PyTorch-based deep learning library provides a framework for analyzing and forecasting temporal data. It implements specialized architectures for time series forecasting, anomaly detection, data imputation, and classification. The project distinguishes itself through the inclusion of zero-shot inference capabilities, allowing large-scale temporal models to be evaluated on unseen datasets without requiring task-specific fine-tuning. The framework covers a broad range of analytical capabilities, including the recovery of missing values in incomplete datasets, the identification of irregul
Implements a framework for evaluating large-scale temporal models on unseen datasets without task-specific fine-tuning.
This project is a multimodal translation framework and large language model capable of speech-to-speech, speech-to-text, and text-to-text translation across nearly 100 languages. It provides a real-time speech translation engine and a comprehensive toolkit for converting spoken audio between languages. The system is distinguished by its ability to preserve the original speaker's tone, pace, and prosody during translation. It utilizes a specialized on-device inference toolkit that converts model checkpoints into C-based libraries, enabling low-latency execution on mobile and edge hardware with
Identifies toxic content in speech by analyzing audio embeddings without requiring language-specific training data.
InternVL is a vision-language model framework that fuses a visual encoder with a large language model to translate image features into textual tokens for reasoning. It provides a system for multimodal inference and dialogue, enabling the processing of images and text to answer questions or generate descriptions. The project is distinguished by its high-resolution image processing, which uses dynamic tiling to maintain detail for images up to 4K resolution, and its chain-of-thought visual reasoning for solving complex mathematical and spatial problems. It also supports temporal frame sampling
Assigns predefined labels to images without task-specific training, supporting multiple languages.
AutoGluon is an automated machine learning framework and multimodal library designed to automate the end-to-end pipeline from data preprocessing to high-accuracy model training and validation. It functions as an automated model trainer for tabular, image, text, and time series data, as well as a tool for time series forecasting and foundation model finetuning. The project is distinguished by its ability to jointly process and fuse different data types, allowing for the construction of multimodal neural networks that integrate images, text, and structured tables. It supports zero-shot inferenc
Categorizes images into previously unseen classes by leveraging pretrained vision-language models.
ImageBind is a multi-modal embedding model and joint representation learner that maps images, text, audio, and other modalities into a single shared vector space. It functions as a cross-modal retrieval framework designed to bind multiple sensory inputs into one cohesive mathematical embedding. The system uses a contrastive learning architecture to align disparate data types by maximizing the similarity between related samples. This allows the model to perform zero-shot multimodal classification and execute cross-modal data retrieval, such as locating visual content via natural language descr
Performs zero-shot classification across different modalities without requiring category-specific training examples.
Skorch es una biblioteca que envuelve redes neuronales de PyTorch en una interfaz compatible con scikit-learn, permitiendo que los modelos de aprendizaje profundo se utilicen dentro de pipelines de machine learning estándar y herramientas de optimización de hiperparámetros. Funciona como un adaptador de datos, gestor de entrenamiento y herramienta de optimización que cierra la brecha entre los módulos de aprendizaje profundo y los flujos de trabajo de machine learning convencionales. El proyecto se distingue por proporcionar un kit de herramientas para automatizar el ciclo de vida de entrenamiento de PyTorch, incluyendo checkpointing integrado, parada temprana y programación de tasas de aprendizaje. Además, permite el aprendizaje por transferencia (transfer learning) mediante utilidades para congelar capas específicas del modelo y ajustar pesos preentrenados para tareas especializadas. La superficie de capacidades se extiende a la transformación de datos, incluyendo la conversión de datos tabulares y arrays numéricos a formatos de tensor y el registro de tokenizadores de texto. También proporciona herramientas para la selección de aceleración por hardware, compilación de módulos just-in-time y modelado de datos probabilísticos para la cuantificación de la incertidumbre. El sistema incluye utilidades para mapear hiperparámetros a argumentos de línea de comandos para garantizar experimentos reproducibles.
Provides capabilities for classifying data using large language models without task-specific training data.
Muzic es una plataforma y framework de deep learning para el análisis, composición y síntesis de música impulsada por IA. Funciona como un framework de generación de música y herramienta de análisis, utilizando modelos de lenguaje grandes y agentes autónomos para orquestar la creación e interpretación de música simbólica y de audio. El proyecto se distingue por sus capacidades multimodales, mapeando el lenguaje natural y la música simbólica en un espacio de incrustación (embedding) conjunto compartido para clasificación zero-shot y recuperación de información. Emplea una variedad de arquitecturas especializadas, incluyendo frameworks de difusión para síntesis de audio, mecanismos de atención de grano dual para consistencia estructural de secuencias largas y un sistema híbrido que combina reglas de teoría musical con redes neuronales. La plataforma cubre una amplia gama de capacidades, incluyendo la generación de secuencias MIDI a partir de texto y letras, síntesis de voz cantada neuronal y transcripción automatizada de letras. También proporciona herramientas para el modelado de estructuras musicales, generación simbólica basada en atributos y la orquestación de herramientas musicales externas a través de agentes autónomos. Las utilidades de soporte incluyen pipelines de ingeniería de datos para la binarización de MIDI a gran escala, codificación de conjuntos de datos y procesamiento de señales de audio para la extracción de notas de melodía y alineación de voz a fonema.
Assigns labels to symbolic music by comparing features against text-based prompt templates without specific training.
Spark NLP es un kit de herramientas para el análisis de texto escalable y aprendizaje automático construido sobre el framework de computación distribuida Apache Spark. Proporciona un framework de aprendizaje automático multimodal y un sistema de tuberías distribuido para secuenciar anotadores para procesar datos lingüísticos a gran escala. La librería incluye un procesador de texto transformer para generar embeddings vectoriales contextuales y un motor de inferencia dedicado para gestionar grandes modelos de lenguaje. El proyecto se distingue por su capacidad para procesar tipos de datos heterogéneos, incluyendo texto, audio e imágenes, dentro de una arquitectura unificada de visión-lenguaje. Admite capacidades avanzadas de IA generativa como prompt engineering, extracción de entidades estructuradas con salida JSON restringida e inferencia local para eliminar la latencia de red. Además, proporciona herramientas para la traducción entre idiomas y la clasificación zero-shot a través de modalidades de texto e imagen. El framework cubre una amplia gama de capacidades, incluyendo el entrenamiento de modelos supervisados para el reconocimiento de entidades y el análisis de sentimientos, así como la respuesta a preguntas extractiva y el resumen de documentos. Integra soporte para bases de datos vectoriales para la búsqueda de similitud y ofrece infraestructura para la aceleración por GPU y la gestión del ciclo de vida del modelo a través de un registro centralizado. El kit de herramientas permite la distribución de modelos y tuberías personalizados a través de un repositorio público y admite el despliegue de modelos mediante APIs REST.
Provides models capable of categorizing text into classes not encountered during training using semantic descriptions.
Nixtla is a time series analysis platform centered on a transformer-based foundation model. It provides zero-shot inference for forecasting and anomaly detection, allowing the system to predict future values for new time series without requiring model retraining. The project is designed for large-scale analysis, using distributed inference scaling and forecast parallelization to process millions of data series. It supports fine-tuning adaptation to adjust pretrained weights for domain-specific datasets and offers deployment options ranging from local execution and private containers to integr
Predicts future values for new time series without requiring retraining by leveraging a pretrained foundation model.
This project is a research framework and toolkit designed for training large-scale vision transformers and multimodal language models. It provides a comprehensive suite for vision-language pretraining, enabling the development of models that map images and text into shared latent spaces. The framework is distinguished by its capabilities in high-fidelity image generation and multimodal research, utilizing normalizing flows and variational autoencoders to produce images from text prompts or class labels. It supports the development of both generative and contrastive models, allowing for a wide
Categorizes images into classes without specific label training by computing embeddings from pretrained models.
LLM Guard is a security firewall and guardrail framework designed to scan and sanitize inputs and outputs for large language models. It functions as a proxy gateway and security layer to block prompt injections, toxicity, and sensitive data leakage while ensuring that model interactions remain compliant with organizational policies. The system distinguishes itself through a modular scanner pipeline that utilizes local model orchestration to eliminate external network dependencies. It supports real-time security filtering via streaming chunk analysis and implements a fail-fast execution model
Uses pre-trained zero-shot classification models to categorize text into sensitive topics without task-specific training.