Classifier

PixelwiseSoftmaxClassifier

class chainercv.links.PixelwiseSoftmaxClassifier(predictor, ignore_label=-1, class_weight=None)

A pixel-wise classifier.

It computes the loss based on a given input/label pair for semantic segmentation.

Parameters

• predictor (Link) – Predictor network.

• ignore_label (int) – A class id that is going to be ignored in evaluation. The default value is -1.

• class_weight (array) – An array that contains constant weights that will be multiplied with the loss values along with the channel dimension. This will be used in chainer.functions.softmax_cross_entropy().

forward(x, t)

Computes the loss value for an image and label pair.

Parameters

• x (Variable) – A variable with a batch of images.

• t (Variable) – A variable with the ground truth image-wise label.

Returns

Loss value.

Return type

Variable

to_cpu()

Copies parameter variables and persistent values to CPU.

This method does not handle non-registered attributes. If some of such attributes must be copied to CPU, the link implementation should override device_resident_accept() to do so.

Returns: self

to_gpu(device=None)

Copies parameter variables and persistent values to GPU.

This method does not handle non-registered attributes. If some of such attributes must be copied to GPU, the link implementation must override device_resident_accept() to do so.

Parameters

device – Target device specifier. If omitted, the current device is used.

Returns: self