---
title: Implementing a Metric
description: >-
  While we strive to include as many metrics as possible in torchmetrics, we
  cannot include them all. We have made it
canonical: 'https://lightning.ai/docs/torchmetrics/stable/user-guide/implement'
---
# Implementing a Metric {#implementing-a-metric}

While we strive to include as many metrics as possible in `torchmetrics`, we cannot include them all. We have made it
easy to implement your own metric, and you can contribute it to `torchmetrics` if you wish. This page will guide
you through the process. If you afterwards are interested in contributing your metric to `torchmetrics`, please
read the [contribution guidelines](/docs/torchmetrics/latest/generated/CONTRIBUTING) and
see this [section](#contributing-metric).

## Base interface {#base-interface}

To implement your own custom metric, subclass the base `Metric` class and implement the following
methods:

- `__init__()`: Each state variable should be called using `self.add_state(...)`.
- `update()`: Any code needed to update the state given any inputs to the metric.
- `compute()`: Computes a final value from the state of the metric.

We provide the remaining interface, such as `reset()` that will make sure to correctly reset all metric
states that have been added using `add_state`. You should therefore not implement `reset()` yourself, only in rare
cases where not all the state variables should be reset to their default value. Adding metric states with `add_state`
will make sure that states are correctly synchronized in distributed settings (DDP). To see how metric states are
synchronized across distributed processes, refer to `add_state` docs from the base
`Metric` class.

Below is a basic implementation of a custom accuracy metric. In the `__init__` method we add the metric states
`correct` and `total`, which will be used to accumulate the number of correct predictions and the total number
of predictions, respectively. In the `update` method we update the metric states based on the inputs to the metric.
Finally, in the `compute` method we compute the final metric value based on the metric states.

```python
from torchmetrics import Metric

class MyAccuracy(Metric):
    def __init__(self, **kwargs):
        super().__init__(**kwargs)
        self.add_state("correct", default=torch.tensor(0), dist_reduce_fx="sum")
        self.add_state("total", default=torch.tensor(0), dist_reduce_fx="sum")

    def update(self, preds: Tensor, target: Tensor) -> None:
        preds, target = self._input_format(preds, target)
        if preds.shape != target.shape:
            raise ValueError("preds and target must have the same shape")

        self.correct += torch.sum(preds == target)
        self.total += target.numel()

    def compute(self) -> Tensor:
        return self.correct.float() / self.total
```

A few important things to note:

* The `dist_reduce_fx` argument to `add_state` is used to specify how the metric states should be reduced between
  batches in distributed settings. In this case we use `"sum"` to sum the metric states across batches. A couple of
  built-in options are available: `"sum"`, `"mean"`, `"cat"`, `"min"` or `"max"`, but a custom reduction is
  also supported.

* In `update` we do not return anything but instead update the metric states in-place.

* In `compute` when running in distributed mode, the states would have been synced before the compute method is
  called. Thus `self.correct` and `self.total` will contain the sum of the metric states across all processes.

## Working with list states {#working-with-list-states}

When initializing metric states with `add_state`, the `default` argument can either be a single tensor (as in the
example above) or an empty list. Most metric will only require a single tensor to accumulate the metric states, but
for some metrics that need access to the individual batch states, it can be useful to use a list of tensors. In the
following example we show how to implement Spearman correlation, which requires access to the individual batch states
because we need to calculate the rank of the predictions and targets.

```python
from torchmetrics import Metric
from torchmetrics.utilities import dim_zero_cat

class MySpearmanCorrCoef(Metric):
    def __init__(self, **kwargs):
        super().__init__(**kwargs)
        self.add_state("preds", default=[], dist_reduce_fx="cat")
        self.add_state("target", default=[], dist_reduce_fx="cat")

    def update(self, preds: Tensor, target: Tensor) -> None:
        self.preds.append(preds)
        self.target.append(target)

    def compute(self):
        # parse inputs
        preds = dim_zero_cat(self.preds)
        target = dim_zero_cat(self.target)
        # some intermediate computation...
        r_preds, r_target = _rank_data(preds), _rank_data(target)
        preds_diff = r_preds - r_preds.mean(0)
        target_diff = r_target - r_target.mean(0)
        cov = (preds_diff * target_diff).mean(0)
        preds_std = torch.sqrt((preds_diff * preds_diff).mean(0))
        target_std = torch.sqrt((target_diff * target_diff).mean(0))
        # finalize the computations
        corrcoef = cov / (preds_std * target_std + eps)
        return torch.clamp(corrcoef, -1.0, 1.0)
```

A few important things to note for this example:

* When working with list states, the `dist_reduce_fx` argument to `add_state` should be set to `"cat"` to
  concatenate the list of tensors across batches.

* When working with list states, The `update(...)` method should append the batch states to the list.

* In the `compute` method the list states behave a bit differently dependeding on whether you are running in
  distributed mode or not. In non-distributed mode the list states will be a list of tensors, while in distributed mode
  the list have already been concatenated into a single tensor. For this reason, we recommend always using the
  `dim_zero_cat` helper function which will standardize the list states to be a single concatenated tensor regardless
  of the mode.

* Calling the `reset` method will clear the list state, deleting any values inserted into it. For this reason, care
  must be taken when referencing list states. If you require the values after your metric is reset, you must first
  copy the attribute to another object (e.g. using `deepcopy.copy`).

## Metric attributes {#metric-attributes}

When done implementing your own metric, there are a few properties and attributes that you may want to set to add
additional functionality. The three attributes to consider are: `is_differentiable`, `higher_is_better` and
`full_state_update`. Note that none of them are strictly required to be set for the metric to work.

```python
from torchmetrics import Metric

class MyMetric(Metric):
    # Set to True if the metric is differentiable else set to False
    is_differentiable: Optional[bool] = None

    # Set to True if the metric reaches it optimal value when the metric is maximized.
    # Set to False if it when the metric is minimized.
    higher_is_better: Optional[bool] = True

    # Set to True if the metric during 'update' requires access to the global metric
    # state for its calculations. If not, setting this to False indicates that all
    # batch states are independent and we will optimize the runtime of 'forward'
    full_state_update: bool = True
```

## Plot interface {#plot-interface}

From torchmetrics v1.0.0 onwards, we also support plotting of metrics through the `.plot()` method. By default this method
will raise `NotImplementedError` but can be implemented by the user to provide a custom plot for the metric.
For any metrics that returns a simple scalar tensor, or a dict of scalar tensors the internal `._plot` method can be
used, that provides the common plotting functionality for most metrics in torchmetrics.

```python
from torchmetrics import Metric
from torchmetrics.utilities.plot import _AX_TYPE, _PLOT_OUT_TYPE

class MyMetric(Metric):
    # set these attributes if you want to use the internal ._plot method
    # bounds are automatically added to the generated plot
    plot_lower_bound: Optional[float] = None
    plot_upper_bound: Optional[float] = None

    def plot(
        self, val: Optional[Union[Tensor, Sequence[Tensor]]] = None, ax: Optional[_AX_TYPE] = None
    ) -> _PLOT_OUT_TYPE:
        return self._plot(val, ax)
```

If the metric returns a more complex output, a custom implementation of the `plot` method is required. For more details
on the plotting API, see the this [page](/docs/torchmetrics/1.9.0/user-guide/plotting.md) .

## Internal implementation details {#internal-implementation-details}

This section briefly describes how metrics work internally. We encourage looking at the source code for more info.
Internally, TorchMetrics wraps the user defined `update()` and `compute()` method. We do this to automatically
synchronize and reduce metric states across multiple devices. More precisely, calling `update()` does the
following internally:

1. Clears computed cache.
2. Calls user-defined `update()`.

Similarly, calling `compute()` does the following internally:

1. Syncs metric states between processes.
2. Reduce gathered metric states.
3. Calls the user defined `compute()` method on the gathered metric states.
4. Cache computed result.

From a user's standpoint this has one important side-effect: computed results are cached. This means that no
matter how many times `compute` is called after one and another, it will continue to return the same result.
The cache is first emptied on the next call to `update`.

`forward` serves the dual purpose of both returning the metric on the current data and updating the internal
metric state for accumulating over multiple batches. The `forward()` method achieves this by combining calls
to `update`, `compute` and `reset`. Depending on the class property `full_state_update`, `forward`
can behave in two ways:

1. If `full_state_update` is `True` or  `full_state_update` is `None` (default) it indicates that the metric during `update` requires access to the full
   metric state and we therefore need to do two calls to `update` to secure that the metric is calculated correctly

   1. Calls `update()` to update the global metric state (for accumulation over multiple batches)
   2. Caches the global state.
   3. Calls `reset()` to clear global metric state.
   4. Calls `update()` to update local metric state.
   5. Calls `compute()` to calculate metric for current batch.
   6. Restores the global state.

2. If `full_state_update` is `False` the metric state of one batch is completely independent of the state
   of other batches, which means that we only need to call `update` once.

   1. Caches the global state.
   2. Calls `reset` the metric to its default state
   3. Calls `update` to update the state with local batch statistics
   4. Calls `compute` to calculate the metric for the current batch
   5. Reduce the global state and batch state into a single state that becomes the new global state

If implementing your own metric, we recommend trying out the metric with `full_state_update` class property set to
both `True` and `False`. If the results are equal, then setting it to `False` will usually give the best
performance.

### `class torchmetrics.Metric(**kwargs)` {#class-torchmetrics-metric-kwargs}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L52-L1181)

Base class for all metrics present in the Metrics API.

This class is inherited by all metrics and implements the following functionality:

1.  Handles the transfer of metric states to the correct device.
    
2.  Handles the synchronization of metric states across processes.
    
3.  Provides properties and methods to control the overall behavior of the metric and its states.
    

The three core methods of the base class are: `add_state()`, `forward()` and `reset()` which should almost never be overwritten by child classes. Instead, the following methods should be overwritten `update()` and `compute()`.

Parameters:

**kwargs** ([`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")) –

additional keyword arguments, see [Advanced metric settings](/docs/torchmetrics/1.9.0/user-guide/overview.md#metric-kwargs) for more info.

*   **compute\_on\_cpu**:
    
    If metric state should be stored on CPU during computations. Only works for list states.
    
*   **dist\_sync\_on\_step**:
    
    If metric state should synchronize on `forward()`. Default is `False`.
    
*   **process\_group**:
    
    The process group on which the synchronization is called. Default is the world.
    
*   **dist\_sync\_fn**:
    
    Function that performs the allgather option on the metric state. Default is a custom implementation that calls `torch.distributed.all_gather` internally.
    
*   **distributed\_available\_fn**:
    
    Function that checks if the distributed backend is available. Defaults to a check of `torch.distributed.is_available()` and `torch.distributed.is_initialized()`.
    
*   **sync\_on\_compute**:
    
    If metric state should synchronize when `compute` is called. Default is `True`.
    
*   **compute\_with\_cache**:
    
    If results from `compute` should be cached. Default is `True`.
    

#### `add_state(name, default, dist_reduce_fx=None, persistent=False)` {#add-state-name-default-dist-reduce-fx-none-persistent-false}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L201-L284)

Add metric state variable. Only used by subclasses.

Metric state variables are either :class:\`~torch.Tensor or an empty list, which can be appended to by the metric. Each state variable must have a unique name associated with it. State variables are accessible as attributes of the metric i.e, if `name` is `"my_state"` then its value can be accessed from an instance `metric` as `metric.my_state`. Metric states behave like buffers and parameters of [`Module`](https://docs.pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module "(in PyTorch v2.15)") as they are also updated when `.to()` is called. Unlike parameters and buffers, metric states are not by default saved in the modules [`state_dict`](https://docs.pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.state_dict "(in PyTorch v2.15)").

Parameters:

*   **name** ([`str`](https://docs.python.org/3/builtins/stdtypes.html#str "(in Python v3.15)")) – The name of the state variable. The variable will then be accessible at `self.name`.
    
*   **default** (`Union`\[[`list`](https://docs.python.org/3/builtins/stdtypes.html#list "(in Python v3.15)"), [`Tensor`](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor "(in PyTorch v2.15)")\]) – Default value of the state; can either be a [`Tensor`](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor "(in PyTorch v2.15)") or an empty list. The state will be reset to this value when `self.reset()` is called.
    
*   **dist\_reduce\_fx** (_Optional_) – Function to reduce state across multiple processes in distributed mode. If value is `"sum"`, `"mean"`, `"cat"`, `"min"` or `"max"` we will use `torch.sum`, `torch.mean`, `torch.cat`, `torch.min` and `` torch.max` `` respectively, each with argument `dim=0`. Note that the `"cat"` reduction only makes sense if the state is a list, and not a tensor. The user can also pass a custom function in this parameter.
    
*   **persistent** (_Optional_) – whether the state will be saved as part of the modules `state_dict`. Default is `False`.
    

Return type:

[`None`](https://docs.python.org/3/builtins/constants.html#None "(in Python v3.15)")

Note

Setting `dist_reduce_fx` to None will return the metric state synchronized across different processes. However, there won’t be any reduction function applied to the synchronized metric state.

The metric states would be synced as follows

*   If the metric state is [`Tensor`](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor "(in PyTorch v2.15)"), the synced value will be a stacked [`Tensor`](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor "(in PyTorch v2.15)") across the process dimension if the metric state was a [`Tensor`](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor "(in PyTorch v2.15)"). The original [`Tensor`](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor "(in PyTorch v2.15)") metric state retains dimension and hence the synchronized output will be of shape `(num_process, ...)`.
    
*   If the metric state is a `list`, the synced value will be a `list` containing the combined elements from all processes.
    

Important

When passing a custom function to `dist_reduce_fx`, expect the synchronized metric state to follow the format discussed in the above note.

Caution

The values inserted into a list state are deleted whenever [`reset()`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric.reset "torchmetrics.Metric.reset") is called. This allows device memory to be automatically reallocated, but may produce unexpected effects when referencing list states. To retain such values after [`reset()`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric.reset "torchmetrics.Metric.reset") is called, you must first copy them to another object.

Raises:

*   [**ValueError**](https://docs.python.org/3/builtins/exceptions.html#ValueError "(in Python v3.15)") – If `default` is not a `tensor` or an `empty list`.
    
*   [**ValueError**](https://docs.python.org/3/builtins/exceptions.html#ValueError "(in Python v3.15)") – If `dist_reduce_fx` is not callable or one of `"mean"`, `"sum"`, `"cat"`, `"min"`, `"max"` or `None`.
    

#### `clone()` {#clone}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L775-L777)

Make a copy of the metric.

Return type:

[`Metric`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric "torchmetrics.metric.Metric")

#### `abstract compute()` {#abstract-compute}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L714-L720)

Override this method to compute the final metric value.

This method will automatically synchronize state variables when running in distributed backend.

Return type:

[`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")

#### `double()` {#double}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L839-L845)

Override default and prevent dtype casting.

Please use [`Metric.set_dtype()`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric.set_dtype "torchmetrics.Metric.set_dtype") instead.

Return type:

[`Metric`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric "torchmetrics.metric.Metric")

#### `float()` {#float}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L831-L837)

Override default and prevent dtype casting.

Please use [`Metric.set_dtype()`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric.set_dtype "torchmetrics.Metric.set_dtype") instead.

Return type:

[`Metric`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric "torchmetrics.metric.Metric")

#### `forward(*args, **kwargs)` {#forward-args-kwargs}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L286-L317)

Aggregate and evaluate batch input directly.

Serves the dual purpose of both computing the metric on the current batch of inputs but also add the batch statistics to the overall accumulating metric state. Input arguments are the exact same as corresponding `update` method. The returned output is the exact same as the output of `compute`.

Parameters:

*   **args** ([`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")) – Any arguments as required by the metric `update` method.
    
*   **kwargs** ([`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")) – Any keyword arguments as required by the metric `update` method.
    

Return type:

[`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")

Returns:

The output of the `compute` method evaluated on the current batch.

Raises:

[**TorchMetricsUserError**](/docs/torchmetrics/1.9.0/api-reference/utilities.md#torchmetrics.utilities.exceptions.TorchMetricsUserError "torchmetrics.utilities.exceptions.TorchMetricsUserError") – If the metric is already synced and `forward` is called again.

#### `half()` {#half}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L847-L853)

Override default and prevent dtype casting.

Please use [`Metric.set_dtype()`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric.set_dtype "torchmetrics.Metric.set_dtype") instead.

Return type:

[`Metric`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric "torchmetrics.metric.Metric")

#### `merge_state(incoming_state)` {#merge-state-incoming-state}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L404-L463)

Merge incoming metric state to the current state of the metric.

Parameters:

**incoming\_state** (`Union`\[[`dict`](https://docs.python.org/3/builtins/stdtypes.html#dict "(in Python v3.15)")\[[`str`](https://docs.python.org/3/builtins/stdtypes.html#str "(in Python v3.15)"), [`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")\], [`Metric`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric "torchmetrics.metric.Metric")\]) – either a dict containing a metric state similar to the metric itself or an instance of the metric class.

Raises:

*   [**ValueError**](https://docs.python.org/3/builtins/exceptions.html#ValueError "(in Python v3.15)") – If the incoming state is neither a dict nor an instance of the metric class.
    
*   [**RuntimeError**](https://docs.python.org/3/builtins/exceptions.html#RuntimeError "(in Python v3.15)") – If the metric has `full_state_update=True` or `dist_sync_on_step=True`. In these cases, the metric cannot be merged with another metric state in a simple way. The user should overwrite the method in the metric class to handle the merge operation.
    
*   [**ValueError**](https://docs.python.org/3/builtins/exceptions.html#ValueError "(in Python v3.15)") – If the incoming state is a metric instance but the class is different from the current metric class.
    

Example with a metric instance: :rtype: [`None`](https://docs.python.org/3/builtins/constants.html#None "(in Python v3.15)")

```python
>>> from torchmetrics.aggregation import SumMetric
>>> metric1 = SumMetric()
>>> metric2 = SumMetric()
>>> metric1.update(1)
>>> metric2.update(2)
>>> metric1.merge_state(metric2)
>>> metric1.compute()
tensor(3.)
```

Example with a dict:

```python
>>> from torchmetrics.aggregation import SumMetric
>>> metric = SumMetric()
>>> metric.update(1)
>>> # SumMetric has one state variable called `sum_value`
>>> metric.merge_state({"sum_value": torch.tensor(2)})
>>> metric.compute()
tensor(3.)
```

#### `persistent(mode=False)` {#persistent-mode-false}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L919-L922)

Change post-init if metric states should be saved to its state\_dict.

Return type:

[`None`](https://docs.python.org/3/builtins/constants.html#None "(in Python v3.15)")

#### `plot(*_, **__)` {#plot}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L722-L724)

Override this method plot the metric value.

Return type:

[`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")

#### `reset()` {#reset}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L758-L773)

Reset metric state variables to their default value.

Return type:

[`None`](https://docs.python.org/3/builtins/constants.html#None "(in Python v3.15)")

#### `set_dtype(dst_type)` {#set-dtype-dst-type}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L855-L865)

Transfer all metric state to specific dtype. Special version of standard type method.

Parameters:

**dst\_type** (`Union`\[[`str`](https://docs.python.org/3/builtins/stdtypes.html#str "(in Python v3.15)"), [`dtype`](https://docs.pytorch.org/docs/stable/tensor_attributes.html#torch.dtype "(in PyTorch v2.15)")\]) – the desired type as string or dtype object

Return type:

[`Metric`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric "torchmetrics.metric.Metric")

#### `state_dict(destination=None, prefix='', keep_vars=False)` {#state-dict-destination-none-prefix-keep-vars-false}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L924-L956)

Get the current state of metric as an dictionary.

Parameters:

*   **destination** ([`Optional`](https://docs.python.org/3/library/typing.html#typing.Optional "(in Python v3.15)")\[[`dict`](https://docs.python.org/3/builtins/stdtypes.html#dict "(in Python v3.15)")\[[`str`](https://docs.python.org/3/builtins/stdtypes.html#str "(in Python v3.15)"), [`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")\]\]) – Optional dictionary, that if provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an `OrderedDict` will be created and returned.
    
*   **prefix** ([`str`](https://docs.python.org/3/builtins/stdtypes.html#str "(in Python v3.15)")) – optional string, a prefix added to parameter and buffer names to compose the keys in state\_dict.
    
*   **keep\_vars** ([`bool`](https://docs.python.org/3/builtins/functions.html#bool "(in Python v3.15)")) – by default the [`Tensor`](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor "(in PyTorch v2.15)") returned in the state dict are detached from autograd. If set to `True`, detaching will not be performed.
    

Return type:

[`dict`](https://docs.python.org/3/builtins/stdtypes.html#dict "(in Python v3.15)")\[[`str`](https://docs.python.org/3/builtins/stdtypes.html#str "(in Python v3.15)"), [`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")\]

#### `sync(dist_sync_fn=None, process_group=None, should_sync=True, distributed_available=None)` {#sync-dist-sync-fn-none-process-group-none-should-sync-true-distributed-available-none}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L573-L615)

Sync function for manually controlling when metrics states should be synced across processes.

Parameters:

*   **dist\_sync\_fn** ([`Optional`](https://docs.python.org/3/library/typing.html#typing.Optional "(in Python v3.15)")\[[`Callable`](https://docs.python.org/3/library/typing.html#typing.Callable "(in Python v3.15)")\]) – Function to be used to perform states synchronization
    
*   **process\_group** ([`Optional`](https://docs.python.org/3/library/typing.html#typing.Optional "(in Python v3.15)")\[[`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")\]) – Specify the process group on which synchronization is called. default: None (which selects the entire world)
    
*   **should\_sync** ([`bool`](https://docs.python.org/3/builtins/functions.html#bool "(in Python v3.15)")) – Whether to apply to state synchronization. This will have an impact only when running in a distributed setting.
    
*   **distributed\_available** ([`Optional`](https://docs.python.org/3/library/typing.html#typing.Optional "(in Python v3.15)")\[[`Callable`](https://docs.python.org/3/library/typing.html#typing.Callable "(in Python v3.15)")\]) – Function to determine if we are running inside a distributed setting
    

Raises:

[**TorchMetricsUserError**](/docs/torchmetrics/1.9.0/api-reference/utilities.md#torchmetrics.utilities.exceptions.TorchMetricsUserError "torchmetrics.utilities.exceptions.TorchMetricsUserError") – If the metric is already synced and `sync` is called again.

Return type:

[`None`](https://docs.python.org/3/builtins/constants.html#None "(in Python v3.15)")

#### `sync_context(dist_sync_fn=None, process_group=None, should_sync=True, should_unsync=True, distributed_available=None)` {#sync-context-dist-sync-fn-none-process-group-none-should-sync-true-should-unsync-true-distributed-available-none}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/../home/runner/.local/share/uv/python/cpython-3.10-linux-x86_64-gnu/lib/python3.10/contextlib.py#L639-L674)

Context manager to synchronize states.

This context manager is used in distributed setting and makes sure that the local cache states are restored after yielding the synchronized state.

Parameters:

*   **dist\_sync\_fn** ([`Optional`](https://docs.python.org/3/library/typing.html#typing.Optional "(in Python v3.15)")\[[`Callable`](https://docs.python.org/3/library/typing.html#typing.Callable "(in Python v3.15)")\]) – Function to be used to perform states synchronization
    
*   **process\_group** ([`Optional`](https://docs.python.org/3/library/typing.html#typing.Optional "(in Python v3.15)")\[[`Any`](https://docs.python.org/3/library/typing.html#typing.Any "(in Python v3.15)")\]) – Specify the process group on which synchronization is called. default: None (which selects the entire world)
    
*   **should\_sync** ([`bool`](https://docs.python.org/3/builtins/functions.html#bool "(in Python v3.15)")) – Whether to apply to state synchronization. This will have an impact only when running in a distributed setting.
    
*   **should\_unsync** ([`bool`](https://docs.python.org/3/builtins/functions.html#bool "(in Python v3.15)")) – Whether to restore the cache state so that the metrics can continue to be accumulated.
    
*   **distributed\_available** ([`Optional`](https://docs.python.org/3/library/typing.html#typing.Optional "(in Python v3.15)")\[[`Callable`](https://docs.python.org/3/library/typing.html#typing.Callable "(in Python v3.15)")\]) – Function to determine if we are running inside a distributed setting
    

Return type:

[`Generator`](https://docs.python.org/3/library/collections.abc.html#collections.abc.Generator "(in Python v3.15)")

#### `type(dst_type)` {#type-dst-type}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L823-L829)

Override default and prevent dtype casting.

Please use [`Metric.set_dtype()`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric.set_dtype "torchmetrics.Metric.set_dtype") instead.

Return type:

[`Metric`](/docs/torchmetrics/1.9.0/api-reference/metric.md#torchmetrics.Metric "torchmetrics.metric.Metric")

#### `unsync(should_unsync=True)` {#unsync-should-unsync-true}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L617-L637)

Unsync function for manually controlling when metrics states should be reverted back to their local states.

Parameters:

**should\_unsync** ([`bool`](https://docs.python.org/3/builtins/functions.html#bool "(in Python v3.15)")) – Whether to perform unsync

Return type:

[`None`](https://docs.python.org/3/builtins/constants.html#None "(in Python v3.15)")

#### `abstract update(*_, **__)` {#abstract-update}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/v1.9.0/src/torchmetrics/metric.py#L710-L712)

Override this method to update the state variables of your metric class.

Return type:

[`None`](https://docs.python.org/3/builtins/constants.html#None "(in Python v3.15)")

#### `property device: device` {#property-device-device}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/torchmetrics.py)

Return the device of the metric.

#### `property dtype: dtype` {#property-dtype-dtype}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/torchmetrics.py)

Return the default dtype of the metric.

#### `property metric_state: dict[str, Union[List[torch.Tensor], torch.Tensor]]` {#property-metric-state-dict-str-union-list-torch-tensor-torch-tensor}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/torchmetrics.py)

Get the current state of the metric.

#### `property update_called: bool` {#property-update-called-bool}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/torchmetrics.py)

Returns True if update or forward has been called initialization or last reset.

#### `property update_count: int` {#property-update-count-int}

[\[source\]](https://github.com/Lightning-AI/torchmetrics/blob/torchmetrics.py)

Get the number of times update and/or forward has been called since initialization or last reset.

## Contributing your metric to TorchMetrics {#contributing-your-metric-to-torchmetrics}

Wanting to contribute the metric you have implemented? Great, we are always open to adding more metrics to `torchmetrics`
as long as they serve a general purpose. However, to keep all our metrics consistent we request that the implementation
and tests gets formatted in the following way:

1. Start by reading our [contribution guidelines](/docs/torchmetrics/latest/generated/CONTRIBUTING).
2. First implement the functional backend. This takes care of all the logic that goes into the metric. The code should
   be put into a single file placed under `src/torchmetrics/functional/"domain"/"new_metric".py` where `domain` is the type of
   metric (classification, regression, text etc.) and `new_metric` is the name of the metric. In this file, there should be the
   following three functions:

  1. `_new_metric_update(...)`: everything that has to do with type/shape checking and all logic required before distributed syncing need to go here.
  2. `_new_metric_compute(...)`: all remaining logic.
  3. `new_metric(...)`: essentially wraps the `_update` and `_compute` private functions into one public function that
     makes up the functional interface for the metric.

> **Hint**
>
> The [functional mean squared error](https://github.com/Lightning-AI/torchmetrics/blob/master/src/torchmetrics/functional/regression/mse.py)
> metric is a is a great example of how to divide the logic.

3. In a corresponding file placed in `src/torchmetrics/"domain"/"new_metric".py` create the module interface:

  1. Create a new module metric by subclassing `torchmetrics.Metric`.
  2. In the `__init__` of the module call `self.add_state` for as many metric states are needed for the metric to
     proper accumulate metric statistics.
  3. The module interface should essentially call the private `_new_metric_update(...)` in its `update` method and similarly the
     `_new_metric_compute(...)` function in its `compute`. No logic should really be implemented in the module interface.
     We do this to not have duplicate code to maintain.

> **Note**
>
> The module [MeanSquaredError](https://github.com/Lightning-AI/torchmetrics/blob/master/src/torchmetrics/regression/mse.py)
> metric that corresponds to the above functional example showcases these steps.

4. Remember to add binding to the different relevant `__init__` files.

5. Testing is key to keeping `torchmetrics` trustworthy. This is why we have a very rigid testing protocol. This means
   that we in most cases require the metric to be tested against some other common framework (`sklearn`, `scipy` etc).

  1. Create a testing file in `tests/unittests/"domain"/test_"new_metric".py`. Only one file is needed as it is intended to test
     both the functional and module interface.
  2. In that file, start by defining a number of test inputs that your metric should be evaluated on.
  3. Create a testclass `class NewMetric(MetricTester)` that inherits from `tests.helpers.testers.MetricTester`.
     This test class should essentially implement the `test_"new_metric"_class` and `test_"new_metric"_fn` methods that
     respectively tests the module interface and the functional interface.
  4. The testclass should be parameterized (using `@pytest.mark.parametrize`) by the different test inputs defined initially.
     Additionally, the `test_"new_metric"_class` method should also be parameterized with an `ddp` parameter such that it gets
     tested in a distributed setting. If your metric has additional parameters, then make sure to also parameterize these
     so that different combinations of inputs and parameters get tested.
  5. (optional) If your metric raises any exception, please add tests that showcase this.

> **Hint**
>
> The [test file for MSE](https://github.com/Lightning-AI/torchmetrics/blob/master/tests/unittests/regression/test_mean_error.py)
> metric shows how to implement such tests.

If you only can figure out part of the steps, do not fear to send a PR. We will much rather receive working
metrics that are not formatted exactly like our codebase, than not receiving any. Formatting can always be applied.
We will gladly guide and/or help implement the remaining :]

---

**Navigation**

- Parent: [User Guide](/docs/torchmetrics/1.9.0/user-guide.md)