model_224614468_albef_nano.py

Model Overview

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

Architecture

• Architecture: albef
• Scale: nano
• Attention: sparse
• Fusion strategy: tensor fusion
• Task head: multitask
• Activation: relu
• Normalization: batchnorm
• Initialization: kaiming normal

Training

• Optimizer: lamb
• LR scheduler: linear warmup

Files

• model_224614468_albef_nano.py — main artifact of this repository

License

See the license field above.