Getatlas Ai6qy6oo3j

Evidently AI

Stay updated on the status of all your company's support tickets.

Customer portal

Evidently AI provides a powerful way to visualize and analyze your model and data monitoring results with interactive reports. These reports help you easily understand your data drift, model performance, and data quality checks, making it easier to communicate insights with your team and stakeholders.

  Ensure you have the necessary packages installed. If not, install them using pip:

Understand how to create interactive reports that visualize data drift, model performance, and data quality.

Generating Interactive Reports

Evidently AI's reports are designed to be informative and insightful, but you can further customize them to fit your specific needs and communicate information effectively to your team and stakeholders.  Here's how to tailor those reports to showcase your most valuable findings.

  Select the metrics that are most important for your model monitoring and data analysis.  You can include metrics from pre-defined presets like `DataDriftPreset`, `DataQualityPreset`, or `RegressionPreset`.  You can also add custom metrics, as we'll see later.

 Control the order in which metrics are displayed in the report. This can be helpful to prioritize certain sections for your audience.

  You can add additional sections to the report to provide further context or information. For example, you could add a section explaining your model, the business context, or specific domain requirements.

  To customize the report, you'll work with the `Report` class. Here's a basic example:

Explore options for customizing report layouts, adding specific visualizations, and exporting reports.

Customizing Report Content

Data monitoring and model performance tracking provide valuable insights that can help improve decision-making and ensure the reliability of your machine learning systems. It's essential to effectively communicate these findings to relevant stakeholders, such as business leaders, product managers, and data scientists, to drive action and build trust in your models. 

 Select the most appropriate way to share your monitoring insights based on your audience and the complexity of the information. Some common options include:

 Prioritize the most important findings and insights from your monitoring reports. Present clear and concise information, highlighting potential issues, trends, and areas requiring attention. 

 Tailor your communication to the audience's level of technical understanding. Avoid jargon and use clear and simple language to ensure everyone understands the message. 

  Explain the significance of your findings by providing context. This might involve clarifying the business impact of data drift, the consequences of model performance degradation, or the implications of data quality issues.

  Leverage data visualizations to present complex information in a clear and concise manner. Visuals can help stakeholders better understand trends, patterns, and relationships within the data. 

  Don't just report problems. Provide clear and actionable recommendations for addressing the identified issues. This might include suggestions for retraining the model, adjusting data preprocessing steps, or addressing data quality problems.

 Create opportunities for questions and discussions to ensure everyone understands the findings and implications of your monitoring reports. 

   Create a clear documentation of your monitoring process, including the metrics used, thresholds for alerts, and procedures for addressing potential issues. This documentation helps maintain transparency and ensures consistency in your reporting processes.

Learn how to effectively share monitoring reports and insights with relevant stakeholders.

Report & Visualization

Generating Interactive Reports

Customizing Report Content

Sharing Monitoring Insights with Stakeholders