metadata
license: bsd-3-clause
tags:
- adamw
- bilinear
- generation
- groupnorm
- kaiming-normal
- linear-warmup
- multi-query
- perceiver
- relu
- tiny
model_082268761_perceiver_tiny.py
Model Overview
A tiny-scale implementation of the perceiver architecture, built for generation tasks.
Architecture
- Architecture: perceiver
- Scale: tiny
- Attention: multi query
- Fusion strategy: bilinear
- Task head: generation
- Activation: relu
- Normalization: groupnorm
- Initialization: kaiming normal
Training
- Optimizer: adamw
- LR scheduler: linear warmup
Files
model_082268761_perceiver_tiny.py— main artifact of this repository
License
See the license field above.