# End-to-end training

This paper proposes a method to reinterpret Vision Transformer (ViT) as an image segmentation model, demonstrating ViT's potential in image segmentation tasks.

This model reveals the potential of Vision Transformer (ViT) in image segmentation tasks by adapting its architecture specifically for segmentation.

The Magician is the first multi-modal large language model with free-form multi-image localization capabilities, achieving precise localization in complex multi-image scenarios and outperforming models with a scale of 70B in performance.

A large language model built from scratch, with fully open-source implementations including tokenizer training, model initialization, pre-training, and instruction fine-tuning

This DETR model has been trained end-to-end on the SKU110K object detection dataset with the number of queries set to 400, suitable for scenarios like product shelf detection.

An image segmentation model fine-tuned on the tontokoton/artery-ultrasound-siit dataset based on nvidia/mit-b0

EnCodec is a high-fidelity real-time neural audio codec developed by Meta AI, employing end-to-end training and supporting multiple bandwidth settings.

Object detection model trained on the LVIS dataset containing 1,203 categories using deformable detection transformer architecture

This is an image classification model generated by HuggingPics, primarily used for classifying images of specific types (genres).

An image classification model generated by HuggingPics for identifying gender (male or female) in images.

YOLOS is an object detection model using Vision Transformer (ViT) architecture, trained with DETR loss and fine-tuned on COCO and Matterport Balloon datasets.

This model is an automatic speech recognition (ASR) model trained on the LibriSpeech ASR dataset, capable of converting English speech into text.

An automatic speech recognition model trained on the LibriSpeech ASR dataset, designed to convert English speech into text.

A lightweight automatic speech recognition (ASR) model, based on a randomly initialized version of the Wav2Vec2 architecture, designed for robustness testing.

This is a FastSpeech2 text-to-speech (TTS) model trained using the ESPnet framework, utilizing the LJSpeech dataset.

This model is a speech recognition model trained on the librispeech_asr dataset, achieving a word error rate (WER) of 2.0133 on the evaluation set.