model_254108194_albef_huge.py
Model Overview
A huge-scale implementation of the albef architecture, built for matching tasks.
Architecture
- Architecture: albef
- Scale: huge
- Attention: multi query
- Fusion strategy: bilinear
- Task head: matching
- Activation: mish
- Normalization: batchnorm
- Initialization: xavier uniform
Training
- Optimizer: sgd
- LR scheduler: polynomial
Files
model_254108194_albef_huge.py — main artifact of this repository
License
See the license field above.