Technology Radar
Published : Oct 26, 2022
NOT ON THE CURRENT EDITION
This blip is not on the current edition of the Radar. If it was on one of the last few editions, it is likely that it is still relevant. If the blip is older, it might no longer be relevant and our assessment might be different today. Unfortunately, we simply don't have the bandwidth to continuously review blips from previous editions of the Radar.
Understand more
Oct 2022
Trial
Feast 是一个用于机器学习的开源 Feature Store 。它具有几个有用的特性，包括生成时间点 (point-in-time) 正确的特征集——以避免在训练时将容易出错的未来特征值泄露给模型——以及支持流数据源与批数据源。然而，它目前仅支持具有时间戳的结构化数据，因此，如果你在模型中处理非结构化数据，那么 Feast 可能并不适用。我们已成功地大规模采用 Feast 作为模型训练时的离线存储与预测时的在线存储。
