model_283929310_coca_xlarge.py

Model Overview

A xlarge-scale implementation of the coca architecture, built for classification tasks.

Architecture

• Architecture: coca
• Scale: xlarge
• Attention: linear
• Fusion strategy: cross attention
• Task head: classification
• Activation: approx gelu
• Normalization: instancenorm
• Initialization: kaiming normal

Training

• Optimizer: sgd
• LR scheduler: cosine

Files

• model_283929310_coca_xlarge.py — main artifact of this repository

License

See the license field above.