public static extern int CvANN_MLPTrain(
IntPtr model,
IntPtr trainData,
IntPtr responses,
IntPtr sampleWeights,
IntPtr sampleIdx,
ref MCvANN_MLP_TrainParams parameters,
MlEnum.ANN_MLP_TRAINING_FLAG flags);
/// <summary>
/// Train the ANN_MLP model with the specific paramters
/// </summary>
/// <param name="trainData">The training data. A 32-bit floating-point, single-channel matrix, one vector per row</param>
/// <param name="responses">A floating-point matrix of the corresponding output vectors, one vector per row. </param>
/// <param name="sampleWeights">It is not null only for RPROP. The optional floating-point vector of weights for each sample. Some samples may be more important than others for training, e.g. user may want to gain the weight of certain classes to find the right balance between hit-rate and false-alarm rate etc</param>
/// <param name="sampleIdx">Can be null if not needed. When specified, identifies samples of interest. It is a Matrix>int< of nx1</param>
/// <param name="parameters">The parameters for ANN_MLP</param>
/// <param name="flag">The traning flag</param>
/// <returns>The number of done iterations</returns>
public int Train(
Matrix<float> trainData,
Matrix<float> responses,
Matrix<float> sampleWeights,
Matrix<int> sampleIdx,
MCvANN_MLP_TrainParams parameters,
MlEnum.ANN_MLP_TRAINING_FLAG flag)
{
return
MlInvoke.CvANN_MLPTrain(
_ptr,
trainData.Ptr,
responses.Ptr,
sampleWeights == null? IntPtr.Zero : sampleWeights.Ptr,
sampleIdx == null ? IntPtr.Zero : sampleIdx.Ptr,
parameters,
flag);
}
/// <summary>
/// Train the decision tree using the specific traning data
/// </summary>
/// <param name="trainData">The training data. A 32-bit floating-point, single-channel matrix, one vector per row</param>
/// <param name="tflag">data layout type</param>
/// <param name="responses">A floating-point matrix of the corresponding output vectors, one vector per row. </param>
/// <param name="varIdx">Can be null if not needed. When specified, identifies variables (features) of interest. It is a Matrix>int< of nx1</param>
/// <param name="sampleIdx">Can be null if not needed. When specified, identifies samples of interest. It is a Matrix>int< of nx1</param>
/// <param name="varType">The types of input variables</param>
/// <param name="missingMask">Can be null if not needed. When specified, it is an 8-bit matrix of the same size as <paramref name="trainData"/>, is used to mark the missed values (non-zero elements of the mask)</param>
/// <param name="param">The parameters for training the decision tree</param>
/// <returns></returns>
public bool Train(
Matrix<float> trainData,
MlEnum.DATA_LAYOUT_TYPE tflag,
Matrix<float> responses,
Matrix<Byte> varIdx,
Matrix<Byte> sampleIdx,
Matrix<Byte> varType,
Matrix<Byte> missingMask,
MCvDTreeParams param)
{
return MlInvoke.CvDTreeTrain(
_ptr,
trainData.Ptr,
tflag,
responses.Ptr,
varIdx == null ? IntPtr.Zero : varIdx.Ptr,
sampleIdx == null ? IntPtr.Zero : sampleIdx.Ptr,
varType == null ? IntPtr.Zero : varType.Ptr,
missingMask == null ? IntPtr.Zero : missingMask.Ptr,
param);
}
/// <summary>
/// Train the boost tree using the specific traning data
/// </summary>
/// <param name="trainData">The training data. A 32-bit floating-point, single-channel matrix, one vector per row</param>
/// <param name="tflag">data layout type</param>
/// <param name="responses">A floating-point matrix of the corresponding output vectors, one vector per row. </param>
/// <param name="varIdx">Can be null if not needed. When specified, identifies variables (features) of interest. It is a Matrix>int< of nx1</param>
/// <param name="sampleIdx">Can be null if not needed. When specified, identifies samples of interest. It is a Matrix>int< of nx1</param>
/// <param name="varType">The types of input variables</param>
/// <param name="missingMask">Can be null if not needed. When specified, it is an 8-bit matrix of the same size as <paramref name="trainData"/>, is used to mark the missed values (non-zero elements of the mask)</param>
/// <param name="param">The parameters for training the boost tree</param>
/// <param name="update">specifies whether the classifier needs to be updated (i.e. the new weak tree classifiers added to the existing ensemble), or the classifier needs to be rebuilt from scratch</param>
/// <returns></returns>
public bool Train(
Matrix<float> trainData,
MlEnum.DATA_LAYOUT_TYPE tflag,
Matrix<float> responses,
Matrix<int> varIdx,
Matrix<int> sampleIdx,
Matrix<int> varType,
Matrix<int> missingMask,
MCvBoostParams param,
bool update)
{
return MlInvoke.CvBoostTrain(
_ptr,
trainData.Ptr,
tflag,
responses.Ptr,
varIdx == null ? IntPtr.Zero : varIdx.Ptr,
sampleIdx == null ? IntPtr.Zero : sampleIdx.Ptr,
varType == null ? IntPtr.Zero : varType.Ptr,
missingMask == null ? IntPtr.Zero : missingMask.Ptr,
param,
update);
}
public static extern void CvSVMGetDefaultGrid(MlEnum.SVM_PARAM_TYPE type, ref MCvParamGrid grid);
public static extern bool CvRTreesTrain(
IntPtr model,
IntPtr trainData,
MlEnum.DATA_LAYOUT_TYPE tFlag,
IntPtr responses,
IntPtr varIdx,
IntPtr sampleIdx,
IntPtr varType,
IntPtr missingMask,
MCvRTParams param);
public static extern IntPtr CvANN_MLPCreate(
IntPtr layerSizes,
MlEnum.ANN_MLP_ACTIVATION_FUNCTION activeFunction,
double fParam1,
double fParam2);
public static extern bool CvBoostTrain(
IntPtr model,
IntPtr trainData,
MlEnum.DATA_LAYOUT_TYPE tFlag,
IntPtr responses,
IntPtr varIdx,
IntPtr sampleIdx,
IntPtr varType,
IntPtr missingMask,
MCvBoostParams param,
[MarshalAs(UnmanagedType.I1)]
bool update);
/// <summary>
/// Get the default parameter grid for the specific SVM type
/// </summary>
/// <param name="type">The SVM type</param>
/// <returns>The default parameter grid for the specific SVM type </returns>
public static MCvParamGrid GetDefaultGrid(MlEnum.SVM_PARAM_TYPE type)
{
MCvParamGrid grid = new MCvParamGrid();
MlInvoke.CvSVMGetDefaultGrid(type, ref grid);
return grid;
}
/// <summary>
/// Create a neural network using the specific parameters
/// </summary>
/// <param name="layerSize">The size of the layer</param>
/// <param name="activeFunction">Activation function</param>
/// <param name="fParam1">Free parameters of the activation function, alpha</param>
/// <param name="fParam2">Free parameters of the activation function, beta</param>
public ANN_MLP(Matrix<int> layerSize, MlEnum.ANN_MLP_ACTIVATION_FUNCTION activeFunction, double fParam1, double fParam2)
{
_ptr = MlInvoke.CvANN_MLPCreate(layerSize.Ptr, activeFunction, fParam1, fParam2);
}
/// <summary>
/// Create an Expectation Maximization model
/// </summary>
/// <param name="nclusters">The number of mixture components in the Gaussian mixture model. Use 5 for default.</param>
/// <param name="covMatType">Constraint on covariance matrices which defines type of matrices</param>
/// <param name="termcrit">The termination criteria of the EM algorithm. The EM algorithm can be terminated by the number of iterations termCrit.maxCount (number of M-steps) or when relative change of likelihood logarithm is less than termCrit.epsilon. Default maximum number of iterations is 100</param>
public EM(int nclusters, MlEnum.EM_COVARIAN_MATRIX_TYPE covMatType, MCvTermCriteria termcrit)
{
_ptr = MlInvoke.CvEMDefaultCreate(nclusters, covMatType, termcrit);
}
public static extern bool CvGBTreesTrain( IntPtr model, IntPtr trainData, MlEnum.DATA_LAYOUT_TYPE tFlag, IntPtr responses, IntPtr varIdx, IntPtr sampleIdx, IntPtr varType, IntPtr missingMask, MCvGBTreesParams param, [MarshalAs(CvInvoke.BoolMarshalType)] bool update);
public static extern IntPtr CvEMDefaultCreate(int nclusters, MlEnum.EM_COVARIAN_MATRIX_TYPE covMatType, MCvTermCriteria termcrit);
