| | --- |
| | datasets: |
| | - Fanqi-Lin/Processed-Task-Dataset |
| | metrics: |
| | - accuracy |
| | pipeline_tag: robotics |
| | --- |
| | # Robotic Manipulation Models for Four Tasks |
| | [[Project Page]](https://data-scaling-laws.github.io/) |
| | [[Paper]](https://data-scaling-laws.github.io/paper.pdf) |
| | [[Code]](https://github.com/Fanqi-Lin/Data-Scaling-Laws) |
| | [[Processed Dataset]](https://huggingface.co/datasets/Fanqi-Lin/Processed-Task-Dataset) |
| | [[Raw GoPro Videos]](https://huggingface.co/datasets/Fanqi-Lin/GoPro-Raw-Videos) |
| |
|
| | This repository contains four models for the manipulation tasks described in the paper "Data Scaling Laws in Imitation Learning for Robotic Manipulation". |
| |
|
| | The tasks include: |
| | + Arrange Mouse |
| | + Fold Towel |
| | + Pour Water |
| | + Unplug Charger |
| |
|
| | For each task, we release a policy trained on data collected from 32 unique environment-object pairs, with 50 demonstrations per environment. These policies have been shown to generalize effectively to novel environments and objects. |
| |
|
| | For details on how to use these models, please refer to our [code](https://github.com/Fanqi-Lin/Data-Scaling-Laws). |
