ares.visualization package

Submodules

ares.visualization.html_header module

ares.visualization.response_visualizer module

Response Visualizer for ARES evaluation results.

Generates HTML visualizations of evaluation results in chat-like format, supporting both multi-turn conversations and single-turn responses.

class ares.visualization.response_visualizer.ResponseVisualizer[source]

Bases: object

Visualizer for ARES evaluation results in chat format.

detect_evaluation_type(results: list[dict[str, Any]]) → str[source]

Detect the type of evaluation file.

Returns:

Individual turn-based evaluations (keyword/prompt) - ‘mixed’: Mix of multi-turn and single-turn evaluations - ‘goal’: Goal-level aggregated evaluations - ‘conversation’: Conversation-level evaluations - ‘single_turn’: Single turn responses without conversation structure

Return type:

  • ‘multi_turn’

extract_conversations_from_goal(results: list[dict[str, Any]]) → list[dict[str, Any]][source]

Extract conversations from goal-level evaluation format.

generate_conversation_html_from_messages(conv_data: dict[str, Any], conv_num: int, total_convs: int) → str[source]

Generate HTML for a conversation from goal/conversation evaluation format.

generate_conversation_html_multi_turn(conv_id: str, turns: list[dict[str, Any]], conv_num: int, total_convs: int) → str[source]

Generate HTML for a multi-turn conversation.

Generate HTML footer.

generate_html_header(title: str = 'ARES Response Viewer') → str[source]

Generate HTML header with CSS styling.

generate_sidebar(items: list[Any], eval_type: str) → str[source]

Generate sidebar navigation.

Note: Uses list.append() + ‘

‘.join() for HTML generation.

If performance becomes an issue with very large outputs, consider using io.StringIO for incremental string building.

generate_single_turn_html(item: dict[str, Any], item_num: int, total_items: int) → str[source]

Generate HTML for a single-turn response.

group_by_conversation(results: list[dict[str, Any]]) → dict[str, list[dict[str, Any]]][source]

Group turns by conversation ID for multi-turn evaluations.

static load_evaluation_file(filepath: Path) → list[dict[str, Any]][source]

Load evaluation JSON file.

static render_markdown(text: str) → str[source]

Convert markdown formatting to HTML.

visualize(filepath: str | Path, output_file: str | Path | None = None, max_items: int | None = None, evaluator_name: str | None = None) → Path[source]

Generate HTML visualization from evaluation file.

Parameters:

  • filepath – Path to evaluation JSON file

  • output_file – Optional output HTML file path

  • max_items – Optional maximum number of items to visualize

  • evaluator_name – Optional name of the evaluator (e.g., ‘keyword’, ‘conversation_eval’)

Returns:

Path to generated HTML file

Module contents

ARES Visualization Module for displaying evaluation results in chat-like format.

class ares.visualization.ResponseVisualizer[source]

Bases: object

Visualizer for ARES evaluation results in chat format.

detect_evaluation_type(results: list[dict[str, Any]]) → str[source]

Detect the type of evaluation file.

Returns:

Individual turn-based evaluations (keyword/prompt) - ‘mixed’: Mix of multi-turn and single-turn evaluations - ‘goal’: Goal-level aggregated evaluations - ‘conversation’: Conversation-level evaluations - ‘single_turn’: Single turn responses without conversation structure

Return type:

  • ‘multi_turn’

extract_conversations_from_goal(results: list[dict[str, Any]]) → list[dict[str, Any]][source]

Extract conversations from goal-level evaluation format.

generate_conversation_html_from_messages(conv_data: dict[str, Any], conv_num: int, total_convs: int) → str[source]

Generate HTML for a conversation from goal/conversation evaluation format.

generate_conversation_html_multi_turn(conv_id: str, turns: list[dict[str, Any]], conv_num: int, total_convs: int) → str[source]

Generate HTML for a multi-turn conversation.

Generate HTML footer.

generate_html_header(title: str = 'ARES Response Viewer') → str[source]

Generate HTML header with CSS styling.

generate_sidebar(items: list[Any], eval_type: str) → str[source]

Generate sidebar navigation.

Note: Uses list.append() + ‘

‘.join() for HTML generation.

If performance becomes an issue with very large outputs, consider using io.StringIO for incremental string building.

generate_single_turn_html(item: dict[str, Any], item_num: int, total_items: int) → str[source]

Generate HTML for a single-turn response.

group_by_conversation(results: list[dict[str, Any]]) → dict[str, list[dict[str, Any]]][source]

Group turns by conversation ID for multi-turn evaluations.

static load_evaluation_file(filepath: Path) → list[dict[str, Any]][source]

Load evaluation JSON file.

static render_markdown(text: str) → str[source]

Convert markdown formatting to HTML.

visualize(filepath: str | Path, output_file: str | Path | None = None, max_items: int | None = None, evaluator_name: str | None = None) → Path[source]

Generate HTML visualization from evaluation file.

Parameters:

  • filepath – Path to evaluation JSON file

  • output_file – Optional output HTML file path

  • max_items – Optional maximum number of items to visualize

  • evaluator_name – Optional name of the evaluator (e.g., ‘keyword’, ‘conversation_eval’)

Returns:

Path to generated HTML file