metadata
license: apache-2.0
tags:
- cosine
- cross-attention
- deit
- groupnorm
- relu
- retrieval
- rmsprop
- sparse
- tiny
- xavier
model_563240713_deit_tiny.py
Model Overview
A tiny-scale implementation of the deit architecture, built for retrieval tasks.
Architecture
- Architecture: deit
- Scale: tiny
- Attention: sparse
- Fusion strategy: cross attention
- Task head: retrieval
- Activation: relu
- Normalization: groupnorm
- Initialization: xavier
Training
- Optimizer: rmsprop
- LR scheduler: cosine
Files
model_563240713_deit_tiny.py— main artifact of this repository
License
See the license field above.