Skip to page content
SAVRN
Search Plan a project

Independent publisher

Harrysunshine

Harrysunshine

Models in Library2
Datasets in Library0
Models on Hugging Face12
Followers—

Compare the published evidence

This profile connects the 2 model and dataset records shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.

Published tasks: Robotics.

Reported licenses: apache-2.0, gemma. Open a record to review its specific license and conditions.

Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.

Models

底座里的 VAE、文本编码器与 Wan2.2-TI2V-5B 扩散主干仍从 Dexmal/DW05-Robotwin 取,本仓只放微调后的权重。 - 起点:Dexmal/DW05-Robotwin,剔掉与双臂机器人绑定的动作头与本体编码器(宽度对不上 SO-101,重新学) - 全量微调 + DeepSpeed ZeRO-1,每卡 batch 1 × 3 卡,20000 步,学习率 1e-4 cosine - SO-101 的 6 维状态填进 RobotWin 的 16 槽排布:前五个槽放五个臂关节,夹爪落在第 7 槽;三路视图用 顶视 + 腕部 + 腕部 - 本体状态在训练时没有归一化:推理时把 6 维状态排进 16 槽、补一位终止标志 0 后原样送入,不要套 RobotWin 的分位数归一化。 用法与评测脚本见 dexbotic 的 docs/so101.md(script/so101/dw05simcheck.py), Apache-2.0,随底座 Dexmal/DW05-Robotwin。

Open weights apache-2.0

在 Dexmal/DM05 上做 LoRA 微调得到的 SO-101 单臂策略: 仿真与真机混成一份,共 3698 集 / 1,280,507 帧 / 10 个任务,全部 30 fps、640×480: - LoRA(all-linear),可训参数 324M / 6.15B = 5.27% - 动作模式 RELATIVE,动作块长度 50 - 等效 batch 48(4 卡 × 每卡 6 × 梯度累积 2),5000 步,bf16,cosine 学习率 成败取环境自带的 issuccess(物体在料箱内、夹爪松开、物体与机械臂都静止),每局上限 500 步: 需要 dexbotic 框架。适配器里记录的底座是 Dexmal/DM05, 完整的数据准备、训练与仿真评测流程见 dexbotic 的 docs/so101.md,教学示例见 适配器随底座 Dexmal/DM05 采用 Gemma 许可。

Open weights gemma peft