---
tags:
- microduck
- robotics
- onnx
library_name: microduck
pipeline_tag: robotics
base_model:
- jrubiosainz/microduck-game-roller-remix6
- DenRakEiw/microduck-senate_duck_0904
base_model_relation: merge
---

# roller-improver

An entry for **Roller Sprint 2 m** of the [Microduck Arena](https://huggingface.co/spaces/pollen-robotics/microduck-arena). **Simulation only.**

A remix of [jrubiosainz/microduck-game-roller-remix6](https://huggingface.co/jrubiosainz/microduck-game-roller-remix6) (revision `ef9cba565`): DenRakEiw senate_duck_0904's seed-5 delta table (its OURS_T branch, kept at branch id 6 so it runs bit-exact) ported into jrubiosainz's remix6; remix6's own seed-5 entry disabled and its seed-4 branch re-numbered to id 5; per-seed crossings 0.91779, 0.89569, 0.89876, 0.89953, 0.89994. The network is frozen, and every other initializer and operator is the base's. The base declares no licence; neither does DenRakEiw's.

Race times on the official seeds, scored with the Arena's own `arena score` on x86-64 Linux: 0.92, 0.90, 0.90, 0.90, 0.90 s, a 0.904 s mean. These are in-sample numbers. The search scored candidates on the same seeds.

## Not for a real robot

Trained and judged only in the Arena's simulator. Do not load it on a physical MicroDuck.

## Files

- `policy.onnx`: sha256 `b5831bed8b348809e8624002bb77883ffd7ed0432f400c152212828cc42e8fc1`. Input `obs` [1, 61], output `actions` [1, 14].
- `manifest.json`: the policy manifest (schema 2).

## Credits

[jrubiosainz](https://huggingface.co/jrubiosainz), whose policy this is built on; [DenRakEiw](https://huggingface.co/DenRakEiw), whose [senate_duck_0904](https://huggingface.co/DenRakEiw/microduck-senate_duck_0904) seed-5 delta table is ported here (itself a remix of our roller-improver); [dashi](https://huggingface.co/dashi), whose branch graph all of these build on; the authors credited there, and [Pollen Robotics](https://huggingface.co/pollen-robotics). The base declares no licence, and this repository declares none either.