Shortcuts

PSNR#

class ignite.metrics.PSNR(data_range, output_transform=<function PSNR.<lambda>>, device=device(type='cpu'), skip_unrolling=False)[source]#

Computes average Peak signal-to-noise ratio (PSNR).

PSNR(I,J)=10log10(MAXI2 MSE )\text{PSNR}(I, J) = 10 * \log_{10}\left(\frac{ MAX_{I}^2 }{ \text{ MSE } }\right)

where MSE\text{MSE} is mean squared error.

  • update must receive output of the form (y_pred, y).

  • y_pred and y must have (batch_size, …) shape.

  • y_pred and y must have same dtype and same shape.

Parameters
  • data_range (Union[int, float]) – The data range of the target image (distance between minimum and maximum possible values). For other data types, please set the data range, otherwise an exception will be raised.

  • output_transform (Callable) – A callable that is used to transform the Engine’s process_function’s output into the form expected by the metric.

  • 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.

  • skip_unrolling (bool) – specifies whether output should be unrolled before being fed to update method. Should be true for multi-output model, for example, if y_pred contains multi-ouput as (y_pred_a, y_pred_b) Alternatively, output_transform can be used to handle this.

Examples

To use with Engine and process_function, simply attach the metric instance to the engine. The output of the engine’s process_function needs to be in format of (y_pred, y) or {'y_pred': y_pred, 'y': y, ...}. If not, output_tranform can be added to the metric to transform the output into the form expected by the metric.

For more information on how metric works with Engine, visit Attach Engine API.

from collections import OrderedDict

import torch
from torch import nn, optim

from ignite.engine import *
from ignite.handlers import *
from ignite.metrics import *
from ignite.metrics.regression import *
from ignite.utils import *

# create default evaluator for doctests

def eval_step(engine, batch):
    return batch

default_evaluator = Engine(eval_step)

# create default optimizer for doctests

param_tensor = torch.zeros([1], requires_grad=True)
default_optimizer = torch.optim.SGD([param_tensor], lr=0.1)

# create default trainer for doctests
# as handlers could be attached to the trainer,
# each test must define his own trainer using `.. testsetup:`

def get_default_trainer():

    def train_step(engine, batch):
        return batch

    return Engine(train_step)

# create default model for doctests

default_model = nn.Sequential(OrderedDict([
    ('base', nn.Linear(4, 2)),
    ('fc', nn.Linear(2, 1))
]))

manual_seed(666)
psnr = PSNR(data_range=1.0)
psnr.attach(default_evaluator, 'psnr')
preds = torch.rand([4, 3, 16, 16])
target = preds * 0.75
state = default_evaluator.run([[preds, target]])
print(state.metrics['psnr'])
16.8671405...

This metric by default accepts Grayscale or RGB images. But if you have YCbCr or YUV images, only Y channel is needed for computing PSNR. And, this can be done with output_transform. For instance,

def get_y_channel(output):
    y_pred, y = output
    # y_pred and y are (B, 3, H, W) and YCbCr or YUV images
    # let's select y channel
    return y_pred[:, 0, ...], y[:, 0, ...]

psnr = PSNR(data_range=219, output_transform=get_y_channel)
psnr.attach(default_evaluator, 'psnr')
preds = 219 * torch.rand([4, 3, 16, 16])
target = preds * 0.75
state = default_evaluator.run([[preds, target]])
print(state.metrics['psnr'])
16.7027966...

New in version 0.4.3.

Changed in version 0.5.1: skip_unrolling argument is added.

Methods

compute

Computes the metric based on its accumulated state.

reset

Resets the metric to its initial state.

update

Updates the metric's state using the passed batch output.

compute()[source]#

Computes the metric based on its 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.

reset()[source]#

Resets the metric to its initial state.

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

Return type

None

update(output)[source]#

Updates the metric’s state using the passed batch output.

By default, this is called once for each batch.

Parameters

output (Sequence[Tensor]) – the is the output from the engine’s process function.

Return type

None