model_470726992_swin_t_xlarge.py

Model Overview

A xlarge-scale implementation of the swin t architecture, built for multitask tasks.

Architecture

• Architecture: swin t
• Scale: xlarge
• Attention: sparse
• Fusion strategy: bilinear
• Task head: multitask
• Activation: approx gelu
• Normalization: rmsnorm
• Initialization: orthogonal

Training

• Optimizer: adam
• LR scheduler: linear warmup

Files

• model_470726992_swin_t_xlarge.py — main artifact of this repository

License

See the license field above.