TrainOcr_SVM
Trains an OCR support vector machines classifier.
Syntax
C++
C#
Python
def TrainOcr_SVM( inCharacterSamples: list[CharacterSample], inCharacterFeatures: CharacterFeatures, outOcrModel: OcrModel, /, *, inNormalizationSize: Size = Size(16, 16), inNu: float | None = None, inKernelGamma: float | None = None, inRegularizationConstant: float = 1.0, inStopEpsilon: float = 0.001, inUseShrinkingHeuristics: bool = True, inCharacterSize: Size | None = None, inRandomSeed: int | None = None ) -> ( outTrainingAccuracy: float, diagNormalizedCharacters: list[Image] )
Parameters
| Name | Type | Range | Default | Description | |
|---|---|---|---|---|---|
![]() |
inCharacterSamples | list[CharacterSample] | Training font created from sample regions | ||
![]() |
inNormalizationSize | Size | Size(16, 16) | The character size after normalization | |
![]() |
inNu | float | None | 0.0 - 1.0 | None | Trade-off between training accuracy and number of supported vectors |
![]() |
inKernelGamma | float | None | None | Gamma parameter for RBF kernel | |
![]() |
inRegularizationConstant | float | 0.0 - ![]() |
1.0 | Preventing overfitting |
![]() |
inStopEpsilon | float | 0.001 | Epsilon for stopping criterion | |
![]() |
inUseShrinkingHeuristics | bool | True | Heuristics may speed up computations | |
![]() |
inCharacterSize | Size | None | None | Size of fixed width font | |
![]() |
inRandomSeed | int | None | 0 - +![]() |
None | Random seed used to train classifier |
![]() |
inCharacterFeatures | CharacterFeatures | Character features used to identify characters | ||
![]() |
outOcrModel | OcrModel | Trained OcrSvmModel used to recognize characters | ||
![]() |
outTrainingAccuracy | float | The overall training score | ||
![]() |
diagNormalizedCharacters | list[Image] | Images of normalized characters used to train classifier |




