model_354426519_mobilevit_large.py

Model Overview

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

Architecture

• Architecture: mobilevit
• Scale: large
• Attention: sparse
• Fusion strategy: gated fusion
• Task head: generation
• Activation: gelu
• Normalization: instancenorm
• Initialization: xavier uniform

Training

• Optimizer: rmsprop
• LR scheduler: onecycle

Files

• model_354426519_mobilevit_large.py — main artifact of this repository

License

See the license field above.