- الوصف :
تم إنشاء مجموعات البيانات هذه باستخدام بيئة PickPlaceCan لمحاكي الذراع الروبوتية robosuite . تم تسجيل مجموعات البيانات البشرية بواسطة مشغل واحد باستخدام RLDS Creator ووحدة تحكم لوحة الألعاب.
تم تسجيل مجموعات البيانات التركيبية باستخدام مكتبة EnvLogger .
تتبع مجموعات البيانات تنسيق RLDS لتمثيل الخطوات والحلقات.
تتكون الحلقات من 400 خطوة. في كل حلقة ، تتم إضافة علامة عند اكتمال المهمة ، ويتم تخزين هذه العلامة كجزء من البيانات الوصفية للخطوة المخصصة.
لاحظ أنه نظرًا لاعتماد EnvLogger ، فإن إنشاء مجموعة البيانات هذه مدعوم حاليًا في بيئات Linux فقط.
كود المصدر :
tfds.rlds.robosuite_panda_pick_place_can.RobosuitePandaPickPlaceCan
إصدارات :
-
1.0.0
(افتراضي): الإصدار الأولي.
-
التخزين المؤقت التلقائي ( التوثيق ): لا
المفاتيح الخاضعة للإشراف (انظر المستند
as_supervised
):None
الشكل ( tfds.show_examples ): غير مدعوم.
الاقتباس :
@misc{ramos2021rlds,
title={RLDS: an Ecosystem to Generate, Share and Use Datasets in Reinforcement Learning},
author={Sabela Ramos and Sertan Girgin and Léonard Hussenot and Damien Vincent and Hanna Yakubovich and Daniel Toyama and Anita Gergely and Piotr Stanczyk and Raphael Marinier and Jeremiah Harmsen and Olivier Pietquin and Nikola Momchev},
year={2021},
eprint={2111.02767},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
robosuite_panda_pick_place_can / human_dc29b40a (التكوين الافتراضي)
وصف التكوين : مجموعة البيانات التي تم إنشاؤها بواسطة الإنسان (50 حلقة).
الصفحة الرئيسية : https://github.com/google-research/rlds
حجم التحميل :
96.67 MiB
حجم مجموعة البيانات :
407.24 MiB
الانقسامات :
انشق، مزق | أمثلة |
---|---|
'train' | 50 |
- هيكل الميزة :
FeaturesDict({
'agent_id': string,
'episode_id': string,
'episode_index': int32,
'steps': Dataset({
'action': Tensor(shape=(7,), dtype=float64),
'discount': float64,
'image': Image(shape=(None, None, 3), dtype=uint8),
'is_first': bool,
'is_last': bool,
'is_terminal': bool,
'observation': FeaturesDict({
'Can_pos': Tensor(shape=(3,), dtype=float64),
'Can_quat': Tensor(shape=(4,), dtype=float64),
'Can_to_robot0_eef_pos': Tensor(shape=(3,), dtype=float64),
'Can_to_robot0_eef_quat': Tensor(shape=(4,), dtype=float32),
'object-state': Tensor(shape=(14,), dtype=float64),
'robot0_eef_pos': Tensor(shape=(3,), dtype=float64),
'robot0_eef_quat': Tensor(shape=(4,), dtype=float64),
'robot0_gripper_qpos': Tensor(shape=(2,), dtype=float64),
'robot0_gripper_qvel': Tensor(shape=(2,), dtype=float64),
'robot0_joint_pos_cos': Tensor(shape=(7,), dtype=float64),
'robot0_joint_pos_sin': Tensor(shape=(7,), dtype=float64),
'robot0_joint_vel': Tensor(shape=(7,), dtype=float64),
'robot0_proprio-state': Tensor(shape=(32,), dtype=float64),
}),
'reward': float64,
'tag:placed': bool,
}),
})
- وثائق الميزة :
ميزة | فصل | شكل | نوع | وصف |
---|---|---|---|---|
الميزات | ||||
معرف_الوكيل | موتر | سلسلة | ||
معرّف_حلقة | موتر | سلسلة | ||
حلقة_الفهرس | موتر | int32 | ||
خطوات | مجموعة البيانات | |||
خطوات / عمل | موتر | (7 ،) | تعويم 64 | |
خطوات / خصم | موتر | تعويم 64 | ||
الخطوات / الصورة | صورة | (لا شيء ، لا شيء ، 3) | uint8 | |
الخطوات / is_first | موتر | منطقي | ||
الخطوات / is_last | موتر | منطقي | ||
الخطوات / is_terminal | موتر | منطقي | ||
خطوات / ملاحظة | الميزات | |||
الخطوات / الملاحظة / Can_pos | موتر | (3 ،) | تعويم 64 | |
خطوات / الملاحظة / Can_quat | موتر | (4 ،) | تعويم 64 | |
الخطوات / الملاحظة / Can_to_robot0_eef_pos | موتر | (3 ،) | تعويم 64 | |
الخطوات / الملاحظة / Can_to_robot0_eef_quat | موتر | (4 ،) | تعويم 32 | |
خطوات / ملاحظة / حالة الكائن | موتر | (14 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_eef_pos | موتر | (3 ،) | تعويم 64 | |
خطوات / الملاحظة / robot0_eef_quat | موتر | (4 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_gripper_qpos | موتر | (2 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_gripper_qvel | موتر | (2 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_joint_pos_cos | موتر | (7 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_joint_pos_sin | موتر | (7 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_joint_vel | موتر | (7 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_proprio-state | موتر | (32 ،) | تعويم 64 | |
خطوات / مكافأة | موتر | تعويم 64 | ||
الخطوات / العلامة: وضعت | موتر | منطقي |
- أمثلة ( tfds.as_dataframe ):
robosuite_panda_pick_place_can / human_images_dc29b40a
وصف التكوين : مجموعة البيانات التي تم إنشاؤها بواسطة الإنسان ، بما في ذلك الصور ذات زوايا الكاميرا المختلفة في الملاحظة. لاحظ أن الأمر قد يستغرق بعض الوقت للإنشاء.
الصفحة الرئيسية : https://www.tensorflow.org/datasets/catalog/robosuite_panda_pick_place_can
حجم التحميل :
10.95 GiB
حجم مجموعة البيانات :
7.53 GiB
الانقسامات :
انشق، مزق | أمثلة |
---|---|
'train' | 50 |
- هيكل الميزة :
FeaturesDict({
'agent_id': string,
'episode_id': string,
'episode_index': int32,
'steps': Dataset({
'action': Tensor(shape=(7,), dtype=float64),
'discount': float64,
'image': Image(shape=(None, None, 3), dtype=uint8),
'is_first': bool,
'is_last': bool,
'is_terminal': bool,
'observation': FeaturesDict({
'Can_pos': Tensor(shape=(3,), dtype=float64),
'Can_quat': Tensor(shape=(4,), dtype=float64),
'Can_to_robot0_eef_pos': Tensor(shape=(3,), dtype=float64),
'Can_to_robot0_eef_quat': Tensor(shape=(4,), dtype=float32),
'agentview_image': Image(shape=(256, 256, 3), dtype=uint8),
'birdview_image': Image(shape=(256, 256, 3), dtype=uint8),
'object-state': Tensor(shape=(14,), dtype=float64),
'robot0_eef_pos': Tensor(shape=(3,), dtype=float64),
'robot0_eef_quat': Tensor(shape=(4,), dtype=float64),
'robot0_eye_in_hand_image': Image(shape=(256, 256, 3), dtype=uint8),
'robot0_gripper_qpos': Tensor(shape=(2,), dtype=float64),
'robot0_gripper_qvel': Tensor(shape=(2,), dtype=float64),
'robot0_joint_pos_cos': Tensor(shape=(7,), dtype=float64),
'robot0_joint_pos_sin': Tensor(shape=(7,), dtype=float64),
'robot0_joint_vel': Tensor(shape=(7,), dtype=float64),
'robot0_proprio-state': Tensor(shape=(32,), dtype=float64),
'robot0_robotview_image': Image(shape=(256, 256, 3), dtype=uint8),
}),
'reward': float64,
'tag:placed': bool,
}),
})
- وثائق الميزة :
ميزة | فصل | شكل | نوع | وصف |
---|---|---|---|---|
الميزات | ||||
معرف_الوكيل | موتر | سلسلة | ||
معرّف_حلقة | موتر | سلسلة | ||
حلقة_الفهرس | موتر | int32 | ||
خطوات | مجموعة البيانات | |||
خطوات / عمل | موتر | (7 ،) | تعويم 64 | |
خطوات / خصم | موتر | تعويم 64 | ||
الخطوات / الصورة | صورة | (لا شيء ، لا شيء ، 3) | uint8 | |
الخطوات / is_first | موتر | منطقي | ||
الخطوات / is_last | موتر | منطقي | ||
الخطوات / is_terminal | موتر | منطقي | ||
خطوات / ملاحظة | الميزات | |||
الخطوات / الملاحظة / Can_pos | موتر | (3 ،) | تعويم 64 | |
خطوات / الملاحظة / Can_quat | موتر | (4 ،) | تعويم 64 | |
الخطوات / الملاحظة / Can_to_robot0_eef_pos | موتر | (3 ،) | تعويم 64 | |
الخطوات / الملاحظة / Can_to_robot0_eef_quat | موتر | (4 ،) | تعويم 32 | |
خطوات / ملاحظة / وكيل view_image | صورة | (256 ، 256 ، 3) | uint8 | |
خطوات / ملاحظة / birdview_image | صورة | (256 ، 256 ، 3) | uint8 | |
خطوات / ملاحظة / حالة الكائن | موتر | (14 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_eef_pos | موتر | (3 ،) | تعويم 64 | |
خطوات / الملاحظة / robot0_eef_quat | موتر | (4 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_eye_in_hand_image | صورة | (256 ، 256 ، 3) | uint8 | |
الخطوات / الملاحظة / robot0_gripper_qpos | موتر | (2 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_gripper_qvel | موتر | (2 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_joint_pos_cos | موتر | (7 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_joint_pos_sin | موتر | (7 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_joint_vel | موتر | (7 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_proprio-state | موتر | (32 ،) | تعويم 64 | |
الخطوات / الملاحظة / robot0_robotview_image | صورة | (256 ، 256 ، 3) | uint8 | |
خطوات / مكافأة | موتر | تعويم 64 | ||
الخطوات / العلامة: وضعت | موتر | منطقي |
- أمثلة ( tfds.as_dataframe ):
robosuite_panda_pick_place_can / الاصطناعية_stochastic_sac_afe13968
وصف التكوين : مجموعة بيانات اصطناعية تم إنشاؤها بواسطة عامل عشوائي تم تدريبه باستخدام SAC (200 حلقة).
الصفحة الرئيسية : https://github.com/google-research/rlds
حجم التحميل :
144.44 MiB
حجم مجموعة البيانات :
622.86 MiB
الانقسامات :
انشق، مزق | أمثلة |
---|---|
'train' | 200 |
- هيكل الميزة :
FeaturesDict({
'agent_id': string,
'episode_id': string,
'episode_index': int32,
'steps': Dataset({
'action': Tensor(shape=(7,), dtype=float32),
'discount': float64,
'image': Image(shape=(None, None, 3), dtype=uint8),
'is_first': bool,
'is_last': bool,
'is_terminal': bool,
'observation': FeaturesDict({
'Can_pos': Tensor(shape=(3,), dtype=float32),
'Can_quat': Tensor(shape=(4,), dtype=float32),
'Can_to_robot0_eef_pos': Tensor(shape=(3,), dtype=float32),
'Can_to_robot0_eef_quat': Tensor(shape=(4,), dtype=float32),
'object-state': Tensor(shape=(14,), dtype=float32),
'robot0_eef_pos': Tensor(shape=(3,), dtype=float32),
'robot0_eef_quat': Tensor(shape=(4,), dtype=float32),
'robot0_gripper_qpos': Tensor(shape=(2,), dtype=float32),
'robot0_gripper_qvel': Tensor(shape=(2,), dtype=float32),
'robot0_joint_pos_cos': Tensor(shape=(7,), dtype=float32),
'robot0_joint_pos_sin': Tensor(shape=(7,), dtype=float32),
'robot0_joint_vel': Tensor(shape=(7,), dtype=float32),
'robot0_proprio-state': Tensor(shape=(32,), dtype=float32),
}),
'reward': float64,
'tag:placed': bool,
}),
})
- وثائق الميزة :
ميزة | فصل | شكل | نوع | وصف |
---|---|---|---|---|
الميزات | ||||
معرف_الوكيل | موتر | سلسلة | ||
معرّف_حلقة | موتر | سلسلة | ||
حلقة_الفهرس | موتر | int32 | ||
خطوات | مجموعة البيانات | |||
خطوات / عمل | موتر | (7 ،) | تعويم 32 | |
خطوات / خصم | موتر | تعويم 64 | ||
الخطوات / الصورة | صورة | (لا شيء ، لا شيء ، 3) | uint8 | |
الخطوات / is_first | موتر | منطقي | ||
الخطوات / is_last | موتر | منطقي | ||
الخطوات / is_terminal | موتر | منطقي | ||
خطوات / ملاحظة | الميزات | |||
الخطوات / الملاحظة / Can_pos | موتر | (3 ،) | تعويم 32 | |
خطوات / الملاحظة / Can_quat | موتر | (4 ،) | تعويم 32 | |
الخطوات / الملاحظة / Can_to_robot0_eef_pos | موتر | (3 ،) | تعويم 32 | |
الخطوات / الملاحظة / Can_to_robot0_eef_quat | موتر | (4 ،) | تعويم 32 | |
خطوات / ملاحظة / حالة الكائن | موتر | (14 ،) | تعويم 32 | |
الخطوات / الملاحظة / robot0_eef_pos | موتر | (3 ،) | تعويم 32 | |
خطوات / الملاحظة / robot0_eef_quat | موتر | (4 ،) | تعويم 32 | |
الخطوات / الملاحظة / robot0_gripper_qpos | موتر | (2 ،) | تعويم 32 | |
الخطوات / الملاحظة / robot0_gripper_qvel | موتر | (2 ،) | تعويم 32 | |
الخطوات / الملاحظة / robot0_joint_pos_cos | موتر | (7 ،) | تعويم 32 | |
الخطوات / الملاحظة / robot0_joint_pos_sin | موتر | (7 ،) | تعويم 32 | |
الخطوات / الملاحظة / robot0_joint_vel | موتر | (7 ،) | تعويم 32 | |
الخطوات / الملاحظة / robot0_proprio-state | موتر | (32 ،) | تعويم 32 | |
خطوات / مكافأة | موتر | تعويم 64 | ||
الخطوات / العلامة: وضعت | موتر | منطقي |
- أمثلة ( tfds.as_dataframe ):