Retrievalmap

---
name: retrievalmap
description: Compute the RetrievalMAP metric — provided by torchmetrics. Use when the user has predictions and ground-truth and needs to compute RetrievalMAP, or asks how to score with RetrievalMAP.
metadata:
  skill_kind: metric
  source_lib: torchmetrics
  import_path: torchmetrics.RetrievalMAP
  source: library_introspection
---

# retrievalmap

> Metric `RetrievalMAP` from `torchmetrics` (torchmetrics.RetrievalMAP)

## When to invoke this skill

The user has predictions + ground truth and asks to evaluate with RetrievalMAP, or
mentions `torchmetrics.RetrievalMAP` directly, or wants the standard torchmetrics implementation.

## Reference signature

```python
from torchmetrics import RetrievalMAP

# _RetrievalMAP(empty_target_action: str = 'neg', ignore_index: Optional[int] = None, top_k: Optional[int] = None, **kwargs: Any) -> None
```

## Library docstring

```
Wrapper for deprecated import.

>>> from torch import tensor
>>> indexes = tensor([0, 0, 0, 1, 1, 1, 1])
>>> preds = tensor([0.2, 0.3, 0.5, 0.1, 0.3, 0.5, 0.2])
>>> target = tensor([False, False, True, False, True, False, True])
>>> rmap = _RetrievalMAP()
>>> rmap(preds, target, indexes=indexes)
tensor(0.7917)
```

## Quick recipe

```python
import torchmetrics as _m
score = _m.RetrievalMAP(y_true, y_pred)
```

## Don'ts

- Don't reimplement when the library version handles edge cases (NaN, ties, empty inputs) better than a hand-rolled formula.
- Always check the library version's argument order — sklearn is `(y_true, y_pred)` while torchmetrics is `(preds, target)`.