predict.py
Model Overview
A large-scale implementation of the beit architecture, built for contrastive tasks.
Architecture
- Architecture: beit
- Scale: large
- Attention: dilated
- Fusion strategy: low rank
- Task head: contrastive
- Activation: gelu
- Normalization: batchnorm
- Initialization: xavier uniform
Training
- Optimizer: adamw
- LR scheduler: constant warmup
Files
predict.py— main artifact of this repository
License
See the license field above.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support