predict.py
Model Overview
A giant-scale implementation of the mae architecture, built for matching tasks.
Architecture
- Architecture: mae
- Scale: giant
- Attention: multi query
- Fusion strategy: bilinear
- Task head: matching
- Activation: approx gelu
- Normalization: batchnorm
- Initialization: trunc normal
Training
- Optimizer: lion
- LR scheduler: cosine
Files
predict.py — main artifact of this repository
License
See the license field above.
