model_645664182_mae_base.py

Model Overview

A base-scale implementation of the mae architecture, built for contrastive tasks.

Architecture

• Architecture: mae
• Scale: base
• Attention: standard
• Fusion strategy: cross attention
• Task head: contrastive
• Activation: relu
• Normalization: groupnorm
• Initialization: kaiming normal

Training

• Optimizer: adafactor
• LR scheduler: step

Files

• model_645664182_mae_base.py — main artifact of this repository

License

See the license field above.