metadata
license: mit
tags:
- cross-attention
- efficientformer
- gelu
- huge
- linear
- matching
- polynomial
- rmsprop
- scalenorm
- xavier-uniform
inference.py
Model Overview
A huge-scale implementation of the efficientformer architecture, built for matching tasks.
Architecture
- Architecture: efficientformer
- Scale: huge
- Attention: linear
- Fusion strategy: cross attention
- Task head: matching
- Activation: gelu
- Normalization: scalenorm
- Initialization: xavier uniform
Training
- Optimizer: rmsprop
- LR scheduler: polynomial
Files
inference.py— main artifact of this repository
License
See the license field above.