model_366767041_mobilevit_huge.py

Model Overview

A huge-scale implementation of the mobilevit architecture, built for generation tasks.

Architecture

• Architecture: mobilevit
• Scale: huge
• Attention: dilated
• Fusion strategy: bilinear
• Task head: generation
• Activation: swish
• Normalization: groupnorm
• Initialization: trunc normal

Training

• Optimizer: lion
• LR scheduler: polynomial

Files

• model_366767041_mobilevit_huge.py — main artifact of this repository

License

See the license field above.