model_563519876_deit_tiny.py

Model Overview

A tiny-scale implementation of the deit architecture, built for generation tasks.

Architecture

• Architecture: deit
• Scale: tiny
• Attention: sparse
• Fusion strategy: tucker
• Task head: generation
• Activation: relu
• Normalization: batchnorm
• Initialization: xavier uniform

Training

• Optimizer: rmsprop
• LR scheduler: exponential

Files

• model_563519876_deit_tiny.py — main artifact of this repository

License

See the license field above.