Classification (Coca)
Overview
A research-oriented Coca prototype targeting Classification. The included nano setup documents defaults and file formats without presenting unverified performance numbers.
Expect rough edges — this is a working sketch, not a polished release.
Limitations
- weights are random-init, so outputs carry no signal
- robustness, fairness and domain transfer are untested
- the API may change without notice while the design settles
Evaluation guidance
- pick a task dataset such as a task-specific labeled split
- fix a tuning budget in advance
- compare against a capacity-matched baseline
- publish raw logs next to any claim
Repository status
One Python module contains the model and a runnable example.
Generated configuration lives in config.json; the recipe in training_args.json.
The checkpoint is an initialization artifact and should not be read as a trained model.
Scores are deliberately absent.
Default experiment recipe
The included configuration uses rmsprop with a exponential schedule. These are starting values in the script, not evidence of a completed run. For a meaningful evaluation, train all baselines with the same data exposure, tuning budget, and random seeds.
Architecture
| Item | Value |
|---|---|
| Architecture | Coca |
| Scale | nano |
| Attention | flash |
| Fusion | low rank |
| Activation | relu |
| Normalization | layernorm |
Quick check
Run python eval.py --help for the built-in smoke test. Because this is a custom implementation, generic automatic loading APIs require an explicit adapter before use.
Changelog
- 2026-10-08 — initial upload
- 2026-10-20 — config and docs sync
Contact
Questions or corrections: open a discussion in this repository.
Files
| File | Purpose |
|---|---|
eval.py |
primary artifact |
README.md |
this documentation |
config.json |
architecture configuration |
training_args.json |
default experiment settings |
model.safetensors |
initialization checkpoint |
USE_POLICY.md |
use and redistribution notes |
License
Distributed under the apache-2.0 license. External datasets keep their own terms — check them separately.
- Downloads last month
- -