Implementation:Evidentlyai Evidently Snapshot Export Methods

Knowledge Sources	Evidently Evidently Report Guide
Domains	ML_Monitoring, Data_Serialization
Last Updated	2026-02-14 12:00 GMT

Overview

Concrete methods for exporting Evidently Snapshot results to HTML, JSON, and Python dict formats provided by the Evidently library.

Description

The Snapshot class provides three export methods:

• save_html(filename): Writes an interactive HTML report with Plotly visualizations
• json(): Returns a JSON string of metrics and test results
• dict(): Returns a Python dictionary with structure {"metrics": [...], "tests": [...]}

Additionally, save_json(filename) writes the JSON to a file.

Usage

Call these methods on a Snapshot returned by Report.run() to persist or transmit evaluation results.

Code Reference

Source Location

• Repository: evidently
• File: src/evidently/core/report.py
• Lines: L639-780

Signature

class Snapshot:
    def dict(self) -> dict:
        """Returns {"metrics": [...], "tests": [...]} simplified format."""

    def json(self) -> str:
        """Returns JSON string of metrics and test results."""

    def save_html(self, filename: Union[str, typing.IO]):
        """Saves interactive HTML report to file."""

    def save_json(self, filename: Union[str, typing.IO]):
        """Saves JSON to file."""

Import

# Snapshot is returned by Report.run(), no direct import needed
from evidently import Report
snapshot = report.run(data)  # Returns Snapshot

I/O Contract

Inputs

Name	Type	Required	Description
filename (save_html)	Union[str, IO]	Yes	Output file path or file-like object
filename (save_json)	Union[str, IO]	Yes	Output file path or file-like object

Outputs

Name	Type	Description
dict() returns	dict	{"metrics": [...], "tests": [...]} with metric values
json() returns	str	JSON string of metrics and test results
save_html()	File	Interactive HTML report written to disk
save_json()	File	JSON written to disk

Usage Examples

Export to HTML

snapshot = report.run(current_data, reference_data)
snapshot.save_html("drift_report.html")

Extract Metric Values

snapshot = report.run(current_data, reference_data)
result = snapshot.dict()

# Access individual metric values
for metric in result["metrics"]:
    print(metric)

# Access test results
for test in result["tests"]:
    print(test)

Store as JSON

import json

snapshot = report.run(current_data, reference_data)

# Get JSON string
json_str = snapshot.json()

# Or save directly to file
snapshot.save_json("results.json")

Related Pages

Implements Principle

• Principle:Evidentlyai_Evidently_Snapshot_Export