- وصف :
تتكون مجموعة البيانات الخاصة بنا من كائنات ذات مظهر وهندسة متنوعة. يتطلب الأمر مهارات حركية دقيقة متعددة المراحل ومتعددة الوسائط لتجميع الأوتاد بنجاح على لوحة غير مثبتة في مشهد عشوائي. لقد جمعنا إجمالي 22550 مسارًا عبر مهمتين مختلفتين على ذراع فرانكا باندا. نسجل المسارات من منظرين عالميين ومنظرين للمعصم. يحتوي كل عرض على كلاً من RGB وخريطة العمق.
الصفحة الرئيسية : https://functional-manipulation-benchmark.github.io/
كود المصدر :
tfds.robotics.rtx.Fmbالإصدارات :
-
0.1.0(افتراضي): الإصدار الأولي.
-
حجم التحميل :
Unknown sizeحجم مجموعة البيانات :
356.63 GiBالتخزين المؤقت التلقائي ( الوثائق ): لا
الإنشقاقات :
| ينقسم | أمثلة |
|---|---|
'train' | 1,804 |
- هيكل الميزة :
FeaturesDict({
'episode_metadata': FeaturesDict({
'episode_language_embedding': Tensor(shape=(512,), dtype=float32),
'episode_language_instruction': string,
'episode_task': string,
'file_path': string,
}),
'steps': Dataset({
'action': Tensor(shape=(7,), dtype=float32),
'discount': Scalar(shape=(), dtype=float32),
'is_first': bool,
'is_last': bool,
'is_terminal': bool,
'language_embedding': Tensor(shape=(512,), dtype=float32),
'language_instruction': string,
'observation': FeaturesDict({
'color_id': Scalar(shape=(), dtype=uint8),
'eef_force': Tensor(shape=(3,), dtype=float32),
'eef_pose': Tensor(shape=(7,), dtype=float32),
'eef_torque': Tensor(shape=(3,), dtype=float32),
'eef_vel': Tensor(shape=(6,), dtype=float32),
'image_side_1': Image(shape=(256, 256, 3), dtype=uint8),
'image_side_1_depth': Tensor(shape=(256, 256), dtype=float32),
'image_side_2': Image(shape=(256, 256, 3), dtype=uint8),
'image_side_2_depth': Tensor(shape=(256, 256), dtype=float32),
'image_wrist_1': Image(shape=(256, 256, 3), dtype=uint8),
'image_wrist_1_depth': Tensor(shape=(256, 256), dtype=float32),
'image_wrist_2': Image(shape=(256, 256, 3), dtype=uint8),
'image_wrist_2_depth': Tensor(shape=(256, 256), dtype=float32),
'joint_pos': Tensor(shape=(7,), dtype=float32),
'joint_vel': Tensor(shape=(7,), dtype=float32),
'length': string,
'object_id': Scalar(shape=(), dtype=uint8),
'primitive': string,
'shape_id': Scalar(shape=(), dtype=uint8),
'size': string,
'state_gripper_pose': Scalar(shape=(), dtype=float32),
}),
'reward': Scalar(shape=(), dtype=float32),
}),
})
- وثائق الميزة :
| ميزة | فصل | شكل | نوع D | وصف |
|---|---|---|---|---|
| المميزاتDict | ||||
| الحلقة_البيانات الوصفية | المميزاتDict | |||
| Episode_metadata/episode_language_embedding | الموتر | (512،) | float32 | |
| Episode_metadata/episode_language_instruction | الموتر | خيط | ||
| Episode_metadata/episode_task | الموتر | خيط | ||
| Episode_metadata/file_path | الموتر | خيط | ||
| خطوات | مجموعة البيانات | |||
| الخطوات/الإجراء | الموتر | (7،) | float32 | |
| الخطوات/الخصم | العددية | float32 | ||
| الخطوات/is_first | الموتر | منطقي | ||
| الخطوات/is_last | الموتر | منطقي | ||
| الخطوات/is_terminal | الموتر | منطقي | ||
| الخطوات/language_embedding | الموتر | (512،) | float32 | |
| الخطوات/language_instruction | الموتر | خيط | ||
| الخطوات/الملاحظة | المميزاتDict | |||
| الخطوات/الملاحظة/color_id | العددية | uint8 | ||
| الخطوات/الملاحظة/eef_force | الموتر | (3،) | float32 | |
| الخطوات/الملاحظة/eef_pose | الموتر | (7،) | float32 | |
| الخطوات/الملاحظة/eef_torque | الموتر | (3،) | float32 | |
| الخطوات/الملاحظة/eef_vel | الموتر | (6،) | float32 | |
| الخطوات/الملاحظة/image_side_1 | صورة | (256، 256، 3) | uint8 | |
| الخطوات/الملاحظة/image_side_1_عمق | الموتر | (256، 256) | float32 | |
| الخطوات/الملاحظة/image_side_2 | صورة | (256، 256، 3) | uint8 | |
| الخطوات/الملاحظة/image_side_2_عمق | الموتر | (256، 256) | float32 | |
| الخطوات/الملاحظة/image_wrist_1 | صورة | (256، 256، 3) | uint8 | |
| الخطوات/الملاحظة/image_wrist_1_عمق | الموتر | (256، 256) | float32 | |
| الخطوات/الملاحظة/image_wrist_2 | صورة | (256، 256، 3) | uint8 | |
| الخطوات/الملاحظة/image_wrist_2_عمق | الموتر | (256، 256) | float32 | |
| الخطوات/الملاحظة/joint_pos | الموتر | (7،) | float32 | |
| الخطوات/الملاحظة/joint_vel | الموتر | (7،) | float32 | |
| الخطوات / الملاحظة / الطول | الموتر | خيط | ||
| الخطوات/الملاحظة/object_id | العددية | uint8 | ||
| الخطوات / الملاحظة / البدائية | الموتر | خيط | ||
| الخطوات/الملاحظة/shape_id | العددية | uint8 | ||
| الخطوات/الملاحظة/الحجم | الموتر | خيط | ||
| الخطوات/الملاحظة/state_gripper_pose | العددية | float32 | ||
| خطوات/مكافأة | العددية | float32 |
المفاتيح الخاضعة للإشراف (راجع
as_superviseddoc ):Noneالشكل ( tfds.show_examples ): غير مدعوم.
أمثلة ( tfds.as_dataframe ):
- الاقتباس :
https://doi.org/10.48550/arXiv.2401.08553