# Attention Mechanism

This is an ONNX port of sentence-transformers/all-MiniLM-L6-v2, adjusted to return attention weights, specifically designed for BM42 search scenarios.

Author style representation model trained on the Reddit Million User Dataset (MUD) for learning universal author representations

A model using Transformer architecture for time series classification, suitable for industrial scenarios such as engine fault detection

DeiT is a data-efficient image Transformer model trained with attention mechanisms, pretrained and fine-tuned on the ImageNet-1k dataset at 224x224 resolution.