model_271955923_coca_small.py

Model Overview

A small-scale implementation of the coca architecture, built for matching tasks.

Architecture

• Architecture: coca
• Scale: small
• Attention: sliding window
• Fusion strategy: gated fusion
• Task head: matching
• Activation: gelu
• Normalization: groupnorm
• Initialization: xavier uniform

Training

• Optimizer: novograd
• LR scheduler: linear warmup

Files

• model_271955923_coca_small.py — main artifact of this repository

License

See the license field above.