CoolFace
Datasetpublic

jellyho/droid_subsets_wiping

DROID Skill Subset: wiping This is a skill-filtered subset of the local DROID 1.0.1 LeRobot-format dataset. It was generated with droid_filter.py from /scratch/jellyho/prsl_skills. Repository Hugging Face repo: jellyho/droid_subsets_wiping Source dataset path used locally: /scratch/jellyho/droid_1.0.1 Output subset path used locally: /scratch/jellyho/droid_subsets/wiping Filtering { "skills": [ "wiping" ], "skill_keywords": {… See the full description on the dataset page: https://huggingface.co/datasets/jellyho/droid_subsets_wiping.

sourceHugging Faceapache-2.0updated 4mo agoView on Hugging Face
0likes118downloads
Dataset Card

DROID Skill Subset: wiping

This is a skill-filtered subset of the local DROID 1.0.1 LeRobot-format dataset. It was generated with droid_filter.py from /scratch/jellyho/prsl_skills.

Repository

  • —Hugging Face repo: jellyho/droid_subsets_wiping
  • —Source dataset path used locally: /scratch/jellyho/droid_1.0.1
  • —Output subset path used locally: /scratch/jellyho/droid_subsets/wiping

Filtering

json
{
  "skills": [
    "wiping"
  ],
  "skill_keywords": {
    "wiping": [
      "wipe",
      "clean",
      "sweep",
      "brush",
      "scrub"
    ]
  },
  "queries": [],
  "exclude": [],
  "regex": false,
  "match_all": false,
  "min_length": null,
  "max_length": null,
  "max_episodes": 300,
  "sample_mode": "balanced",
  "seed": 0
}

Dataset Statistics

  • —Episodes: 300
  • —Frames: 118845
  • —Unique tasks: 300
  • —Sampling mode: balanced
  • —Max episodes: 300
  • —Seed: 0
  • —Mean episode length: 396.15
  • —Min/max episode length: 73 / 1457
  • —Episode success rate: 1.0

Preserved Action Fields

Filtering only selects episodes/rows. It does not drop action-space columns. The following action columns are preserved from the original DROID parquet files:

text
action.gripper_position
action.gripper_velocity
action.cartesian_position
action.cartesian_velocity
action.joint_position
action.joint_velocity
action.original
action

action is the compact 8D policy action used by LeRobot-style loaders. The additional action.* columns remain available in the parquet files for inspection or custom loaders.

Top Tasks

task_textepisodes
Use the towel to wipe the bottle1
Sweep the candy to the left using the white and black brush1
Use the paper towel to wipe the table1
Wipe the table with the blue cloth1
Use the sponge to wipe the red bowl1
Move the white mug to the left then use the green towel to wipe the countertop.1
Use the white towel to wipe the table then put the black towel on the table1
Use the white towel to wipe the table surface1
Clean the table using the yellow towel1
Wipe the striped bowl with the paper towel1

Files

This repository keeps the LeRobot-style layout:

text
meta/info.json
meta/tasks.parquet
meta/episodes/chunk-000/file-000.parquet
data/chunk-000/file-*.parquet
episodes.csv
episodes.jsonl
summary.json
stats.json
task_counts.csv
length_histogram.csv
source_episode_files.csv

The subset keeps compact episode_index, task_index, and index values for loader compatibility. episodes.csv keeps original_episode_index for tracing rows back to the original DROID dataset when available.

Notes

Video files are not included unless explicitly materialized with video symlinks or copies. These subsets are intended for lightweight skill-conditioned fine-tuning experiments.