SAN JOSE, Calif., Sept. 14, 2023 (GLOBE NEWSWIRE) -- Aporia, the leading Observability platform for AI products, today announced a partnership with Snowflake, the Data Cloud company, designed to enhance performance by providing AI models and data observability in the cloud. Integrating Aporia's ML observability platform with the Data Cloud not only promotes the efficiency and effectiveness of data science, but also helps to expand the adoption of responsible AI.

加利福尼亚州圣何塞，2023年9月14日（GLOBE NEWSWIRE）-领先的AI产品可观察平台Aporia今天宣布与数据云公司雪花合作，旨在通过提供AI模型和数据来提高性能。云中的可观察性。将Aporia的ML可观察性平台与数据云集成不仅可以提高数据科学的效率和有效性，还有助于扩大负责任AI的采用。

Aporia’s observability platform now seamlessly integrates with Snowflake, enabling users to monitor their data and production machine learning (ML) models at scale. The integration extends to features within Snowflake such as Snowpark, the runtimes and libraries that make deployment and processing of non-SQL code in Snowflake simpler, faster, and more secure.

Aporia的可观察性平台现在与雪花无缝集成，使用户能够大规模监控他们的数据和生产机器学习（ML）模型。集成扩展到雪花中的功能，例如雪花，运行时间和库，这些功能使雪花中非SQL代码的部署和处理更简单，更快，更安全。

Together, joint customers can monitor billions of predictions to uncover insights from production data and improve their ML models' performance. In addition, Aporia’s privacy-first mode ensures that no sensitive data leaves a user’s Snowflake environment. Aporia supports all AI/ML use cases and every model type including Tabular, NLP, LLM, and Computer Vision within the Snowflake platform.

联合客户可以共同监控数十亿次预测，从生产数据中发现见解并提高其ML模型的性能。此外，Aporia的隐私优先模式可确保没有敏感数据留下用户的雪花环境。Aporia支持雪花平台内的所有AI/ML用例和每种型号，包括表格，NLP，LLM和计算机视觉。

In sensitive domains, gaining insight into the reasons behind specific predictions made by AI projects and the ML models driving them is critical. However, the lack of explainability can hinder the identification of biases, the assurance of fairness, and establishing trust in models. Aporia’s integration with Snowflake addresses these concerns by enabling users to monitor ML models within minutes.

在敏感领域，深入了解AI项目做出的特定预测背后的原因以及驱动它们的ML模型至关重要。然而，缺乏可解释性可能会妨碍偏见的识别，公平的保证以及对模型的信任。Aporia与雪花的集成通过使用户能够在几分钟内监控ML模型来解决这些问题。

It facilitates near real-time detection for issues like AI hallucinations, drift, bias, data integrity issues, model decay, and performance degradation. Aporia's platform cent.

它有助于对AI幻觉，漂移，偏差，数据完整性问题，模型衰减和性能下降等问题进行近实时检测。Aporia的平台中心。