CBinaryCrossEntropyLossLayer Class

June 14, 2020 ยท View on GitHub

This class implements a layer that calculates a cross-entropy loss function for binary classification.

The function is calculated according to the formula:

loss = y * -log(sigmoid(x)) + (1 - y) * -log(1 - sigmoid(x))

where:

  • x is the network response.
  • y is the correct class label (can be 1 or -1).

Please note that this function first calculates a sigmoid on the network response. It is best not to connect this layer input to the output of another sigmoid-calculating layer.

Settings

Correct classification weight

void SetPositiveWeight( float value );

Sets the multiplier for the term that corresponds to the objects for which the class has been detected correctly. You can tune this value to prioritize precision (set value < 1) or recall (set value > 1) during training.

The default value is 1.

Loss weight

void SetLossWeight( float lossWeight );

Sets the multiplier for this function gradient during training. The default value is 1. You may wish to change the default if you are using several loss functions in your network.

Gradient clipping

void SetMaxGradientValue( float maxValue );

Sets the upper limit for the absolute value of the function gradient. Whenever the gradient exceeds this limit its absolute value will be reduced to GetMaxGradientValue().

Trainable parameters

This layer has no trainable parameters.

Inputs

The layer may have 2 to 3 inputs:

  1. The network output for which you are calculating the loss function. Height, Width, Depth, and Channels dimensions of this blob should be equal to 1.
  2. A blob of the same size as the first input, containing the class labels (may be -1 or 1).
  3. [Optional] The objects' weights. This blob should have the same dimensions as the first input.

Outputs

This layer has no output.

Getting the value of the loss function

float GetLastLoss() const;

Use this method to get the value of the loss function calculated on the network's last run.