model_239452297_albef_base.py

Model Overview

A base-scale implementation of the albef architecture, built for matching tasks.

Architecture

• Architecture: albef
• Scale: base
• Attention: sliding window
• Fusion strategy: gated fusion
• Task head: matching
• Activation: approx gelu
• Normalization: groupnorm
• Initialization: kaiming normal

Training

• Optimizer: rmsprop
• LR scheduler: cosine

Files

• model_239452297_albef_base.py — main artifact of this repository

License

See the license field above.