model_558536720_deit_nano.py
Model Overview
A nano-scale implementation of the deit architecture, built for multitask tasks.
Architecture
- Architecture: deit
- Scale: nano
- Attention: linear
- Fusion strategy: low rank
- Task head: multitask
- Activation: relu
- Normalization: instancenorm
- Initialization: kaiming normal
Training
- Optimizer: adamw
- LR scheduler: cosine
Files
model_558536720_deit_nano.py — main artifact of this repository
License
See the license field above.