model_225001889_albef_nano.py

Model Overview

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

Architecture

• Architecture: albef
• Scale: nano
• Attention: flash
• Fusion strategy: tucker
• Task head: matching
• Activation: swish
• Normalization: batchnorm
• Initialization: trunc normal

Training

• Optimizer: rmsprop
• LR scheduler: cosine

Files

• model_225001889_albef_nano.py — main artifact of this repository

License

See the license field above.