Principle:Evidentlyai Evidently Snapshot Export
| Knowledge Sources | |
|---|---|
| Domains | ML_Monitoring, Data_Serialization |
| Last Updated | 2026-02-14 12:00 GMT |
Overview
A result serialization mechanism that exports computed evaluation results into multiple portable formats.
Description
Snapshot Export is the process of converting a Snapshot object (the output of Report.run()) into consumable formats for downstream use. Evidently supports three export formats:
- HTML: Interactive visual report with plots and tables (for human review)
- JSON: Structured text format (for programmatic processing and storage)
- Python dict: Native dictionary with metrics and test results (for in-process use)
The export step bridges evaluation results with external systems (databases, dashboards, file storage, APIs).
Usage
Use this principle after running a Report to persist or transmit results. Choose HTML for visual review, JSON for storage/transmission, and dict for in-process metric extraction (e.g., inserting specific values into a database).
Theoretical Basis
Snapshot export follows the serialization pattern where an in-memory object graph is converted to portable formats:
# Pseudocode: Multi-format serialization
snapshot = report.run(data)
# Format 1: Visual (human-readable)
snapshot.save_html("report.html")
# Format 2: Structured text (machine-readable)
json_str = snapshot.json()
# Format 3: Native object (in-process)
result_dict = snapshot.dict()
# Access: result_dict["metrics"][0]["value"]
Related Pages
Implemented By
Page Connections
Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment