> ## Documentation Index
> Fetch the complete documentation index at: https://fpde-80-mintlify-48090872.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Interpreting results

> Read FPDE signs, evidence values, and result metadata correctly

An FPDE explanation is a target-versus-rival feature attribution.
Read every value in that local contrast.

## Attribution signs

Positive attribution values support the target class.
Negative attribution values support the rival class.
Values near zero contribute little to that local contrast.

```python theme={null}
for name, value in zip(feature_names, attributions):
    if value > 0:
        side = "supports the target class"
    elif value < 0:
        side = "supports the rival class"
    else:
        side = "is neutral in this contrast"
    print(name, value, side)
```

<Note>
  The sign is relative to the selected target class and rival class.
  A feature can change sign under a different rival class.
</Note>

## Evidence and exactness

`details["evidence"]` is the scalar contrast decomposed by the attribution vector.
For an exact decomposition, the attribution sum should be close to the evidence value.

```python theme={null}
print(attributions.sum())
print(details["evidence"])
print(details["exactness_residual"])
```

Small residuals are expected because of floating-point arithmetic.
Large residuals suggest a numerical or input-shape problem.

## Target and rival labels

With `FPDEEngine`, the target class comes from the highest `predict_proba` value.
The rival class comes from the second-highest `predict_proba` value.

Use these labels when you explain results to readers.
Avoid saying that an attribution explains every class at once.

## Raw and normalized values

Use raw attributions when you report evidence-scale values.
Use normalized attributions only for display or comparison.

```python theme={null}
from fpde import diff_fpde

explanation = diff_fpde(x, target_prototype, rival_prototype)
print(explanation.attributions)
print(explanation.normalized_attributions)
```

If you call lower-level functions and receive an `FPDEExplanation`, use `normalized_attributions` for display-scale summaries.
Keep `attributions` for the raw decomposition.

## What not to claim

* Do not describe FPDE as a causal explanation method.
* Do not claim support for prototype types beyond class-mean prototypes in `fpde 0.1.0`.
* Do not compare attribution signs across samples without checking each sample's target and rival class.