Loading…
Loading…
This is an experimental ViT codebase for Matching. It keeps the base setup intentionally manageable so architecture changes can be inspected before a full training run.
config.json records the generated architecture settings.training_args.json records the default experiment recipe.model.safetensors is a valid initialization checkpoint for smoke tests; it is not presented as a trained benchmark checkpoint.| Item | Value |
|---|---|
| Architecture | ViT |
| Scale | base |
| Attention | linear |
| Fusion | co attention |
| Activation | gelu |
| Normalization | layernorm |
The included configuration uses rmsprop with a onecycle 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.
python eval.py --help
Inspect the script's __main__ block for its generated smoke-test example. Because this is a custom implementation, generic automatic loading APIs require an explicit adapter before use.
A useful first evaluation would use a paired validation set, report the task metric across at least three seeds, and include a matched-capacity baseline. Keep training logs and environment versions with any published result.
The initialization checkpoint has not been trained or audited for robustness, fairness, or domain transfer. The implementation should be treated as an experimental starting point. Results from a future trained checkpoint must be documented separately from the defaults shipped here.
eval.py — primary artifactREADME.md — this documentationconfig.json — architecture configurationtraining_args.json — default experiment settingsmodel.safetensors — initialization checkpointReleased under apache-2.0. Review the source-data terms separately when this repository is used with external datasets.