Skip to content
Other

Arm Playground

PPO-политики для 4-суставной руки

1 min read

Arm Playground — набор из двух PPO-политик для симулированной 4-суставной руки (yaw, shoulder, elbow, wrist) на PyTorch. reach наводит кисть на 3D-цель и перехватывает движущуюся, push закатывает шайбу в цель. Политика каждые 20 мс читает состояние и выдаёт 4 команды скорости суставов — без inverse kinematics и планировщика.

Ключевые особенности

  • Две задачи: reach с 21-мерным наблюдением и push с 25-мерным.
  • Выход — 4 числа в [-1, 1], масштабируются vmax = [2.0, 2.0, 2.5, 3.0] rad/s.
  • Форматы .safetensors и .onnx; нормализация и клиппинг уже внутри политики.
  • Обучение на CPU: PPO с 1024 параллельными руками; push — сначала DAgger, затем PPO.
  • Есть live demo в браузере; паритет PyTorch/ONNX проверен до ~1e-6.

Пример

import numpy as np
import onnxruntime as ort
from huggingface_hub import hf_hub_download
path = hf_hub_download("zerochance1907/arm-playground", "reach.onnx")
policy = ort.InferenceSession(path)
obs = np.zeros((1, 21), dtype=np.float32)
action = policy.run(["action"], {"obs": obs})[0]
print(action)  # [[yaw, shoulder, elbow, wrist]] в [-1, 1]
ПодробнееReach: 94% попаданий в 3 см, медиана 0.68 с, ошибка 1.1 см.
Push: 99.5% доставлено, 97.5% остаётся в цели, медиана 1.1 с.
Крошечная MLP 128→128, запускается на CPU.
Ограничения: только собственный симулятор, без sim-to-real и шума; не для реальной руки.
Discuss Telegram 0 19

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # zerochance1907/arm-playground
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download zerochance1907/arm-playground --local-dir ~/models/arm-playground
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to arm-playground
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of arm-playground (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

zerochance1907/arm-playground
Author
zerochance1907
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

190

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments