AOS55/FetchPickAndPlace-v4_sac_100000

这个数据集包含了多个与机器人或智能体相关的观测、动作和奖励等信息，例如实现的目标、期望的目标、当前观测等，都是浮点数序列。还包括了剧集ID、步骤ID、动作、奖励、是否终止、是否成功等字段。此外，还有关于数据集元数据的字段，如算法名称、平均奖励、收集日期、环境名称和剧集数量。数据集分为训练集，大约有10000000个样本，大小为约3.75GB。

This dataset includes multiple observations, actions, and rewards related to robots or agents, such as achieved goals, desired goals, current observations, etc., all of which are sequences of floats. It also includes fields such as episode IDs, step IDs, actions, rewards, terminals, is_success. In addition, there are metadata fields about the dataset, such as algorithm name, average reward, collection date, environment name, and number of episodes. The dataset is split into a training set with about 10 million samples, totaling about 3.75GB in size.