model_448679478_swin_t_nano.py

Model Overview

A nano-scale implementation of the swin t architecture, built for matching tasks.

Architecture

• Architecture: swin t
• Scale: nano
• Attention: sliding window
• Fusion strategy: tensor fusion
• Task head: matching
• Activation: relu
• Normalization: groupnorm
• Initialization: xavier

Training

• Optimizer: adafactor
• LR scheduler: step

Files

• model_448679478_swin_t_nano.py — main artifact of this repository

License

See the license field above.