Abstract class representing a loss function. More...
#include <loss.h>
Inheritance diagram for Loss:
Public Member Functions | |
| Loss () | |
| Constructor for the Loss class. More... | |
| float | calculateMetric (const sTensor &targets, const sTensor &predictions) override |
| Defines the method inherited from abstract Metric parent. More... | |
| virtual sTensor | calculateLoss (const sTensor &targets, const sTensor &predictions)=0 |
 Public Member Functions inherited from Metric | |
| Metric () | |
| Constructor for the Metric class. More... | |
| void | reset () |
Reset the metric, i.e.: set numSamples and currentTotalMetric to 0. | |
| virtual std::string | getShortName () const =0 |
| Short string identifier of the metric. More... | |
Additional Inherited Members | |
| Protected Attributes inherited from Metric | |
| size_t | numSamples |
| The number of samples processed so far. | |
| float | currentTotalMetric |
| The current total value of the metric. | |
Detailed Description
Abstract class representing a loss function.
All loss functions are by default also metrics.
The child classes need to define the calculateLoss and calculateDerivatives methods.
Constructor & Destructor Documentation
◆ Loss()
| Loss::Loss | ( | ) |
Constructor for the Loss class.
Initializes the numSamples and currentTotalLoss variables to 0.
Member Function Documentation
◆ calculateMetric()
|
overridevirtual |
Defines the method inherited from abstract Metric parent.
It simply calls the #calculateLoss function, which is implemented by every child loss function.
- Parameters
-
targets The desired outputs of the network. predictions The actual outputs of the network.
- Returns
- The value of the metric. Here, the value of the loss function.
Implements Metric.
The documentation for this class was generated from the following files:
