Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
passed: bool
task_success: bool
eligible_for_training: bool
started_utc: string
config: struct<purpose: string, synchronous_capture: bool, manual_success_check: bool, task: string, robot: (... 1168 chars omitted)
child 0, purpose: string
child 1, synchronous_capture: bool
child 2, manual_success_check: bool
child 3, task: string
child 4, robot: string
child 5, action_mode: string
child 6, layout: string
child 7, seed: int64
child 8, num_envs: int64
child 9, robot_scale: double
child 10, sources: list<item: string>
child 0, item: string
child 11, floor_contact_paths: list<item: null>
child 0, item: null
child 12, fixed_objects: struct<obj: string, distr_counter: string, distr_sink: string>
child 0, obj: string
child 1, distr_counter: string
child 2, distr_sink: string
child 13, warmup_steps: int64
child 14, camera_data_types: list<item: string>
child 0, item: string
child 15, camera_update_period: double
child 16, planning: struct<position_bounds: struct<mobilebase_forward: list<item: double>, mobilebase_side: list<item: d (... 388 chars omitted)
child 0, position_bounds: struct<mobilebase_forward: list<item: double>, mobilebase_side: list<item: double>, mobilebase_yaw: (... 19 chars omitted)
child 0, mobilebase_forward: list<item: double>
child 0, item: double
child 1, mobilebase_side: list<item: double>
child 0, item: double
child 2, mobileb
...
ble
episode_sha256: string
dataset_validation: struct<transitions: int64, observations: int64, control_dt: double, passed: bool, lift_height_m: dou (... 177 chars omitted)
child 0, transitions: int64
child 1, observations: int64
child 2, control_dt: double
child 3, passed: bool
child 4, lift_height_m: double
child 5, payload_relative_translation_drift_m: double
child 6, gripper_contact_max_n: struct<left_gripper_contact: double, right_gripper_contact: double>
child 0, left_gripper_contact: double
child 1, right_gripper_contact: double
child 7, stable_final_success_steps: int64
observation_storage: string
source_episode: string
source_episode_sha256: string
initial_state_group: string
video_sha256: null
transitions_sha256: string
validation: struct<transitions: int64, observations: int64, control_dt: double, passed: bool, lift_height_m: dou (... 177 chars omitted)
child 0, transitions: int64
child 1, observations: int64
child 2, control_dt: double
child 3, passed: bool
child 4, lift_height_m: double
child 5, payload_relative_translation_drift_m: double
child 6, gripper_contact_max_n: struct<left_gripper_contact: double, right_gripper_contact: double>
child 0, left_gripper_contact: double
child 1, right_gripper_contact: double
child 7, stable_final_success_steps: int64
initial_joints: list<item: double>
child 0, item: double
initial_object_pose: list<item: double>
child 0, item: double
report_sha256: string
video: null
to
{'passed': Value('bool'), 'report_sha256': Value('string'), 'episode_sha256': Value('string'), 'video_sha256': Value('null'), 'transitions_sha256': Value('string'), 'initial_state_group': Value('string'), 'initial_object_pose': List(Value('float64')), 'initial_joints': List(Value('float64')), 'validation': {'transitions': Value('int64'), 'observations': Value('int64'), 'control_dt': Value('float64'), 'passed': Value('bool'), 'lift_height_m': Value('float64'), 'payload_relative_translation_drift_m': Value('float64'), 'gripper_contact_max_n': {'left_gripper_contact': Value('float64'), 'right_gripper_contact': Value('float64')}, 'stable_final_success_steps': Value('int64')}, 'video': Value('null')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 149, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 129, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 544, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 3052, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2559, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2584, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 558, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 440, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 361, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2381, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2309, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
passed: bool
task_success: bool
eligible_for_training: bool
started_utc: string
config: struct<purpose: string, synchronous_capture: bool, manual_success_check: bool, task: string, robot: (... 1168 chars omitted)
child 0, purpose: string
child 1, synchronous_capture: bool
child 2, manual_success_check: bool
child 3, task: string
child 4, robot: string
child 5, action_mode: string
child 6, layout: string
child 7, seed: int64
child 8, num_envs: int64
child 9, robot_scale: double
child 10, sources: list<item: string>
child 0, item: string
child 11, floor_contact_paths: list<item: null>
child 0, item: null
child 12, fixed_objects: struct<obj: string, distr_counter: string, distr_sink: string>
child 0, obj: string
child 1, distr_counter: string
child 2, distr_sink: string
child 13, warmup_steps: int64
child 14, camera_data_types: list<item: string>
child 0, item: string
child 15, camera_update_period: double
child 16, planning: struct<position_bounds: struct<mobilebase_forward: list<item: double>, mobilebase_side: list<item: d (... 388 chars omitted)
child 0, position_bounds: struct<mobilebase_forward: list<item: double>, mobilebase_side: list<item: double>, mobilebase_yaw: (... 19 chars omitted)
child 0, mobilebase_forward: list<item: double>
child 0, item: double
child 1, mobilebase_side: list<item: double>
child 0, item: double
child 2, mobileb
...
ble
episode_sha256: string
dataset_validation: struct<transitions: int64, observations: int64, control_dt: double, passed: bool, lift_height_m: dou (... 177 chars omitted)
child 0, transitions: int64
child 1, observations: int64
child 2, control_dt: double
child 3, passed: bool
child 4, lift_height_m: double
child 5, payload_relative_translation_drift_m: double
child 6, gripper_contact_max_n: struct<left_gripper_contact: double, right_gripper_contact: double>
child 0, left_gripper_contact: double
child 1, right_gripper_contact: double
child 7, stable_final_success_steps: int64
observation_storage: string
source_episode: string
source_episode_sha256: string
initial_state_group: string
video_sha256: null
transitions_sha256: string
validation: struct<transitions: int64, observations: int64, control_dt: double, passed: bool, lift_height_m: dou (... 177 chars omitted)
child 0, transitions: int64
child 1, observations: int64
child 2, control_dt: double
child 3, passed: bool
child 4, lift_height_m: double
child 5, payload_relative_translation_drift_m: double
child 6, gripper_contact_max_n: struct<left_gripper_contact: double, right_gripper_contact: double>
child 0, left_gripper_contact: double
child 1, right_gripper_contact: double
child 7, stable_final_success_steps: int64
initial_joints: list<item: double>
child 0, item: double
initial_object_pose: list<item: double>
child 0, item: double
report_sha256: string
video: null
to
{'passed': Value('bool'), 'report_sha256': Value('string'), 'episode_sha256': Value('string'), 'video_sha256': Value('null'), 'transitions_sha256': Value('string'), 'initial_state_group': Value('string'), 'initial_object_pose': List(Value('float64')), 'initial_joints': List(Value('float64')), 'validation': {'transitions': Value('int64'), 'observations': Value('int64'), 'control_dt': Value('float64'), 'passed': Value('bool'), 'lift_height_m': Value('float64'), 'payload_relative_translation_drift_m': Value('float64'), 'gripper_contact_max_n': {'left_gripper_contact': Value('float64'), 'right_gripper_contact': Value('float64')}, 'stable_final_success_steps': Value('int64')}, 'video': Value('null')}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
CoRe Mobile Manipulation — Trajectory Optimization v1
本数据集为认知推理课程项目的轨迹优化示范，使用课程独立任务进行物理仿真、同步点云和动作采集，并通过逐条独立数据检查。实验室 CCMoMa pipeline 仅作为批量规划实现参考，实验室数据未混入本数据集。
数据规模与任务
| 项目 | 本次快照 |
|---|---|
| 成功示范 | 225 episodes |
| 初始种子 | 23 |
| 训练 / 验证 / 测试 | 205 / 0 / 20 |
| 控制步数 | 532384 |
| 总仿真时长 | 10647.68 s |
| 控制周期 | 0.02 s (50 Hz) |
| NPZ 总大小 | 7.50 GB |
- Robot: PandaOmron-Rel；task: PnPCounterToSink；layout: robocasakitchen-4-2。
- Target: Apricot008，从台面抓起、实际抬升和搬运，释放到水槽，夹爪退开后保持成功。
- Method: cuRobo 全身轨迹优化，复用同一初始场景中已经成功物理执行的抓取前段，批量生成后续搬运候选。
- 全部示范均重新经过完整物理执行与观测采集；抓取和搬运通过位置驱动及物理接触完成。状态写入仅用于执行开始前恢复初始场景。
- This is a partial, growing dataset. The production target of 3000 demonstrations is not yet complete. Raw planned candidates and failed executions are excluded.
文件组织
episodes/train/episode-0001-seed-0001/
episode.npz
report.json
config.yaml
audit.json
episodes/test/...
metadata/fixed-seed-plan.json
metadata/dependencies.lock.toml
metadata/course-source.tar.gz
manifest.json
SHA256SUMS
每条 NPZ 包含 13 维实测关节状态、13 维实际下发的绝对关节位置目标、1024×6 XYZ/RGB 点云，以及位姿、接触力、阶段和成功标志。按本轮存储约定，不保存三路完整 RGB-D、相机标定参数或外部录像；点云包含训练需要的视觉观测。模型、场景资产和训练权重不包含在此数据集中。
读取与 DP3 对齐
import numpy as np
with np.load("episodes/train/episode-0001-seed-0001/episode.npz", allow_pickle=False) as e:
state = e["observation"][:-1] # [H, 13]
action = e["action"] # [H, 13]
point_cloud = e["point_cloud"][:-1] # [H, 1024, 6]
joint_names = e["joint_names"]
joint_units = e["joint_units"]
原始状态和点云有 H+1 帧，动作有 H 帧；对应关系为 observation[k] → action[k] → observation[k+1]，不要将终末观测与下一条轨迹动作配对。点云坐标系为固定 articulation 根坐标系，XYZ 单位为米，RGB 为 0–1。关节名与单位按 NPZ 中字段解释，底盘与夹指为米、旋转关节为弧度。
复现代码快照中的 scripts/export_dp3.py 可分别将 train 和 test 转换成官方 DP3 Zarr，布局为 data/state、data/action、data/point_cloud、meta/episode_ends。使用的是关节位置动作；不能直接当作末端位姿或关节增量动作。
划分与使用边界
按 episode 和初始状态隔离训练与测试；不要随机将相同初始状态的重复示范拆到不同集合。本快照尚无验证集示范，不能将预留测试集用于训练、调参或拟合归一化器。此任务仅覆盖固定机器人、物体和厨房布局；成功示范条数不等同于独立初始状态数，也不是泛化成功率。
任务成功沿用原仿真条件，另外检查实际抬升、携物稳定性、双指接触、末段稳定成功和观测/动作对齐。逐条质量指标见 audit.json 和 report.json。失败样本未包含在成功示范中；全部生产尝试的成功率需结合生产日志另行报告。
解压/下载后可执行 sha256sum -c SHA256SUMS 核对本次快照文件。
来源与复现
- Course repository: https://github.com/IzmKonata/core-mobile-manipulation
- Lab pipeline reference: https://github.com/rlmoma/ccmoma-bench
- Project document: https://ncntrvj1vuzx.feishu.cn/wiki/YESgwauqZiN3epkSDAlcwf8knpd
- Course requirements: https://yzhu.io/courses/core/projects/manipulation/
代码版本、源码哈希和数据哈希见 manifest.json；固定依赖版本与本次课程代码快照见 metadata/。生成数据时包含课程工作区适配，不能仅用 main 的提交号代替完整快照。源代码快照不含凭证、依赖缓存或任何实验室数据；仿真资产按上游安装说明单独准备。
- Downloads last month
- -