# Signal Evaluation

**Module:** `sovai.extensions.signal_evaluation`

## Classes

### `SignalEvaluator`

```python
class SignalEvaluator
```

**Attributes**

* `verbose`
* `df_factor`
* `positions`
* `rebalance_mask`
* `holdings`
* `returns`
* `position_returns`
* `resampled_returns`
* `portfolio_returns`
* `cumulative_returns`
* `performance_plot`
* `performance_table`
* `stress_plot`
* `distribution_plot`
* `drawdown_plot`
* `drawdown_table`
* `returns_heatmap_plot`
* `turnover_plot`
* `signal_correlation_plot`
* `signal_decile_plot`

***

## Functions

### `enrich_figure_with_llm_metadata()`

```python
def enrich_figure_with_llm_metadata(
    fig,
    title: Optional[str] = None,
    description: Optional[str] = None,
    key_stats: Optional[Dict[str, Any]] = None,
    downsample: bool = False,
    target_points: int = 768,
    include_image_export: bool = True,
    image_format: str = 'png',
    image_width: int = 1600,
    image_height: int = 900,
    image_scale: int = 2,
    category: str = 'signal_evaluation',
) -> None
```

Enrich a Plotly figure with LLM-ready metadata.

This function adds metadata to fig.layout.meta that can be used for feeding the figure to an LLM. Modifies the figure in-place.

**Parameters**

| Parameter              | Type                       | Description                                                    |
| ---------------------- | -------------------------- | -------------------------------------------------------------- |
| `fig`                  | —                          | plotly.graph\_objects.Figure The Plotly figure to enrich       |
| `title`                | `Optional[str]`            | str, optional Chart title override                             |
| `description`          | `Optional[str]`            | str, optional Chart description                                |
| `key_stats`            | `Optional[Dict[str, Any]]` | dict, optional Dictionary of key statistics                    |
| `downsample`           | `bool`                     | bool Whether to apply LTTB downsampling in the compressed spec |
| `target_points`        | `int`                      | int Target number of points for downsampling                   |
| `include_image_export` | `bool`                     | bool Whether to generate image export metadata                 |
| `image_format`         | `str`                      | str Image format ('png', 'webp', 'jpeg', 'svg')                |
| `image_width`          | `int`                      | int Image width in pixels                                      |
| `image_height`         | `int`                      | int Image height in pixels                                     |
| `image_scale`          | `int`                      | int Image scale factor                                         |

**Returns**

None (modifies fig in-place)

***

### `export_figure_for_llm()`

```python
def export_figure_for_llm(
    fig,
    include_image: bool = True,
    image_format: str = 'png',
    return_base64: bool = True,
) -> Dict[str, Any]
```

Export a figure with LLM metadata for direct use in LLM prompts.

**Parameters**

| Parameter       | Type   | Description                                                      |
| --------------- | ------ | ---------------------------------------------------------------- |
| `fig`           | —      | plotly.graph\_objects.Figure The Plotly figure with LLM metadata |
| `include_image` | `bool` | bool Whether to export the actual image                          |
| `image_format`  | `str`  | str Image format ('png', 'webp', 'jpeg', 'svg')                  |
| `return_base64` | `bool` | bool Whether to return image as base64 (vs. raw bytes)           |

**Returns**

dict : Contains chart\_card, spec\_capsule, alt\_text, and optionally image

***


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.sov.ai/api-reference/extensions/signal-evaluation.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.