Average

class ignite.metrics.Average(output_transform=<function Average.<lambda>>, device=device(type='cpu'))

Helper class to compute arithmetic average of a single variable.

• update must receive output of the form x.

• x can be a number or torch.Tensor.

Note

Number of samples is updated following the rule:

• +1 if input is a number

• +1 if input is a 1D torch.Tensor

• +batch_size if input is an ND torch.Tensor. Batch size is the first dimension (shape[0]).

For input x being an ND torch.Tensor with N > 1, the first dimension is seen as the number of samples and is summed up and added to the accumulator: accumulator += x.sum(dim=0)

output_tranform can be added to the metric to transform the output into the form expected by the metric.

Parameters

• output_transform (Callable) – a callable that is used to transform the Engine’s process_function’s output into the form expected by the metric. This can be useful if, for example, you have a multi-output model and you want to compute the metric with respect to one of the outputs.

• device (Union[str, device]) – specifies which device updates are accumulated on. Setting the metric’s device to be the same as your update arguments ensures the update method is non-blocking. By default, CPU.

Examples

metric = Average()
metric.attach(default_evaluator, 'avg')
# Case 1. input is er
data = torch.Tensor([0, 1, 2, 3, 4])
state = default_evaluator.run(data)
print(state.metrics['avg'])

2.0

metric = Average()
metric.attach(default_evaluator, 'avg')
# Case 2. input is a 1D torch.Tensor
data = torch.Tensor([
    [0, 0, 0],
    [1, 1, 1],
    [2, 2, 2],
    [3, 3, 3]
])
state = default_evaluator.run(data)
print(state.metrics['avg'])

tensor([1.5000, 1.5000, 1.5000], dtype=torch.float64)

metric = Average()
metric.attach(default_evaluator, 'avg')
# Case 3. input is a ND torch.Tensor
data = [
    torch.Tensor([[0, 0, 0], [1, 1, 1]]),
    torch.Tensor([[2, 2, 2], [3, 3, 3]])
]
state = default_evaluator.run(data)
print(state.metrics['avg'])

tensor([1.5000, 1.5000, 1.5000], dtype=torch.float64)

Methods

compute

Computes the metric based on it's accumulated state.

compute()

Computes the metric based on it’s accumulated state.

By default, this is called at the end of each epoch.

Returns

the actual quantity of interest. However, if a Mapping is returned, it will be (shallow) flattened into engine.state.metrics when completed() is called.

Return type

Any

Raises

NotComputableError – raised when the metric cannot be computed.