List of all members.
Constructor & Destructor Documentation
Constructor that initializes the configuration of the model. Width and Height are set to 1 pixel.
Constructor that initializes the configuration of the model.
- Parameters:
-
imgWidth | Image width in pixels. |
imgHeight | Image height in pixels. |
MAPSTEPS | Maximum number of steps to perform MAP estimation. |
Member Function Documentation
Approximates the expected feature functions using Gibbs Chains.
- Parameters:
-
Input | Input sample. |
Output | Correct output sample used to initialize the Gibbs chain. |
SampleIndex | The current training sample index. |
- Returns:
- A pseudo partition function (which is not to be used for further calculations) and the feature functions (under the models pdf). The forrmat is: Ret[0] = Partition Function, Ret[1] = expected feature function 0, ..., Ret[paramCount + 1] = value of expected feature function paramCount.
void CRFModel_CD::EstimateModelParameters |
( |
const std::vector< double_double_vector > & |
InputSamples, |
|
|
const std::vector< int_int_vector > & |
OutputSamples, |
|
|
size_t |
Steps, |
|
|
double |
StepFactor |
|
) |
| [virtual] |
Estimates the model parameters using contrastive divergence.
- Parameters:
-
InputSamples | Array of input samples. The format is InputSamples[i] = Input Sample i. |
OutputSamples | Array of correctly classified output samples. The format is OutputSamples[i] = Output Sample i. |
Steps | Number of steps to perform steepest decent before terminating. |
StepFactor | Factor determining the step length (in the direction of the loglikelihood gradient) of each iteration. Should be greater than zero and likely less than one. (The actual step length taken is multiplied by the preprocessor variable FeatureFunctionScaleDivision, which was divided by in the function EvaluateNodeVector to improve numerical accuracy on the partition function. This is just a computational practicality and will not affect any other calculations.). |
Reimplemented from CRF::CRFModel.
Sets the parameters of the contrastive divergence learning / inference methods.
- Parameters:
-
ChainSteps | Number of iterations each chain takes. |
ChainCount | Number of chains to average over. |
Member Data Documentation
Number of chains to average over.
Number of iterations each chain takes.
The current test number. Used for experiments.
The documentation for this class was generated from the following files: