model_050432893_cnn_transformer_small.py

Model Overview

A small-scale implementation of the cnn transformer architecture, built for multitask tasks.

Architecture

• Architecture: cnn transformer
• Scale: small
• Attention: grouped query
• Fusion strategy: tensor fusion
• Task head: multitask
• Activation: swish
• Normalization: scalenorm
• Initialization: xavier

Training

• Optimizer: lion
• LR scheduler: cosine

Files

• model_050432893_cnn_transformer_small.py — main artifact of this repository

License

See the license field above.