docs: add detailed dataset card and attribution

README.md

@@ -1,152 +1,76 @@
 ---
 license: other
-task_categories:
-- robotics
 tags:
-- LeRobot
 - robotics
 - lerobot
-- umi
+- manipulation
+- unknown
 - nmotion-initiative
-configs:
-- config_name: default
-  data_files: data/*/*.parquet
 ---
 
-This dataset was created using [LeRobot](https://github.com/huggingface/lerobot).
-
-
-<a class="flex" href="https://huggingface.co/spaces/lerobot/visualize_dataset?path=DaivdYuan/hub-flip-bagel-lerobot">
-<img class="block dark:hidden" src="https://huggingface.co/datasets/huggingface/badges/resolve/main/visualize-this-dataset-xl.svg"/>
-<img class="hidden dark:block" src="https://huggingface.co/datasets/huggingface/badges/resolve/main/visualize-this-dataset-xl-dark.svg"/>
-</a>
+# Flip bagel
 
+## Overview
+This dataset is a conversion from an upstream robotics dataset into **LeRobot v3-compatible format**.
+It is intended to provide reproducible access in a unified schema.
 
-## Dataset Description
+## Source Dataset
+- Dataset ID: `hub-flip-bagel`
+- Project: `unknown`
+- Task: `Flip bagel`
+- Upstream download/source URL: https://real.stanford.edu/maniwav/data/flip/replay_buffer.zarr.zip
+- Upstream project page: https://umi-data.github.io/
+- Contact: liuzeyi@stanford.edu
+- Original paper: Data Scaling Laws in Imitation Learning for Robotic Manipulation (2024)
+- Notes: 283 demos; 1 envs; obs=Image,Proprio,Audio; actions=6 DoF End Effector,Parallel Gripper
 
+## Dataset Format
+- Storage format: LeRobot v3 parquet-based dataset
+- Total episodes: 30
+- Total frames: 9677
+- FPS: 30
+- Robot type: generic_robot
+- Observation feature: `observation.state` with shape `[7]`
+- Action feature: `action` with shape `[7]`
+- Video streams: present
 
+## Video Modalities
+- `observation.images.camera0_rgb`: shape `[224, 224, 3]`
+- Frame rate: `30` FPS
 
-- **Homepage:** [More Information Needed]
-- **Paper:** [More Information Needed]
-- **License:** other
+## Conversion Notes
+- Target schema: LeRobot v3
+- Conversion path: ingest -> preprocess -> zarr probe -> convert 1 episode -> upload
+- `action` fallback may be used when upstream `data/action` is missing.
 
 ## Dataset Structure
-
-[meta/info.json](meta/info.json):
+`meta/info.json`:
 ```json
 {
-    "codebase_version": "v3.0",
-    "robot_type": "generic_robot",
-    "total_episodes": 30,
-    "total_frames": 9677,
-    "total_tasks": 1,
-    "chunks_size": 1000,
-    "data_files_size_in_mb": 100,
-    "video_files_size_in_mb": 200,
-    "fps": 30,
-    "splits": {
-        "train": "0:30"
-    },
-    "data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
-    "video_path": "videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4",
-    "features": {
-        "observation.state": {
-            "dtype": "float32",
-            "shape": [
-                7
-            ],
-            "names": [
-                "state_0",
-                "state_1",
-                "state_2",
-                "state_3",
-                "state_4",
-                "state_5",
-                "state_6"
-            ]
-        },
-        "action": {
-            "dtype": "float32",
-            "shape": [
-                7
-            ],
-            "names": [
-                "action_0",
-                "action_1",
-                "action_2",
-                "action_3",
-                "action_4",
-                "action_5",
-                "action_6"
-            ]
-        },
-        "observation.images.camera0_rgb": {
-            "dtype": "video",
-            "shape": [
-                224,
-                224,
-                3
-            ],
-            "names": [
-                "height",
-                "width",
-                "channels"
-            ],
-            "info": {
-                "video.height": 224,
-                "video.width": 224,
-                "video.codec": "h264",
-                "video.pix_fmt": "yuv420p",
-                "video.is_depth_map": false,
-                "video.fps": 30,
-                "video.channels": 3,
-                "has_audio": false
-            }
-        },
-        "timestamp": {
-            "dtype": "float32",
-            "shape": [
-                1
-            ],
-            "names": null
-        },
-        "frame_index": {
-            "dtype": "int64",
-            "shape": [
-                1
-            ],
-            "names": null
-        },
-        "episode_index": {
-            "dtype": "int64",
-            "shape": [
-                1
-            ],
-            "names": null
-        },
-        "index": {
-            "dtype": "int64",
-            "shape": [
-                1
-            ],
-            "names": null
-        },
-        "task_index": {
-            "dtype": "int64",
-            "shape": [
-                1
-            ],
-            "names": null
-        }
-    }
+  "codebase_version": "v3.0",
+  "robot_type": "generic_robot",
+  "total_episodes": 30,
+  "total_frames": 9677,
+  "fps": 30,
+  "splits": {"train": "0:30"},
+  "data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
+  "video_path": "videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4",
+  "features": {
+    "observation.state": {"dtype": "float32", "shape": [7]},
+    "action": {"dtype": "float32", "shape": [7]}
+  }
 }
 ```
 
+## Data Splits
+- `{'train': '0:30'}`
 
 ## Citation
+This dataset was created using [LeRobot](https://github.com/huggingface/lerobot).
+Please also cite the original upstream project and paper listed above.
 
-**BibTeX:**
-
-```bibtex
-[More Information Needed]
-```
+## Attribution
+- Original dataset credit remains with upstream creators.
+- Conversion and curation contribution:
+  - **Convertor: David Yuan, davidy02@stanford.edu, Stanford University**
+  - **NMotion Initiative** (opensource community): *we make embodied data accessible*