| Customize Help
| Save Settings

MclassCopyResult



Function Map
Synopsis
Copy data from a classification result buffer to a classifier or dataset context.
Syntax
void MclassCopyResult(
MIL_ID SrcResultClassId, //in
MIL_INT64 SrcIndex, //in
MIL_ID DstContextClassId, //in
MIL_INT64 DstIndex, //in
MIL_INT64 CopyType, //in
MIL_INT64 ControlFlag //in
)
Description

This function copies data from a classification result buffer to a classifier or dataset context. For example, you can copy the trained classifier result to a classifier context. The specified result buffer from which to copy (the source) must be for the same type of classifier as the specified context or dataset in which to copy (the destination).

You can only copy trained results after calling MclassTrain(). Once you copy the trained results into a classifier context, you can continue to train that context (MclassTrain()) or you can predict with it (MclassPredict()).

Note that this function reference has not been updated for a MIL system added during a MIL update. Refer to the MIL system's release note to see if there is complementary information.
Parameters
This function is not supported on the selected boards.
Parameters
SrcResultClassId

Specifies the identifier of the source classification result buffer from which to copy. The result buffer must have been allocated using MclassAllocResult(). The type of result buffer that you can specify depends on the copy operation (CopyType). Set this parameter to one of the following values.

function map For specifying the source result
Click to summarizeValue Description
Click to summarize ResultBufTrainId

Specifies the identifier of a classification result buffer (M_TRAIN_CNN_RESULT or M_TRAIN_TREE_ENSEMBLE_RESULT) that holds the results produced from calling MclassTrain() with a CNN training context or tree ensemble training context.

SrcIndex

Reserved for future expansion and must be set to M_DEFAULT.

DstContextClassId

Specifies the identifier of the destination classifier or dataset context in which to copy. The classifier or dataset context must have been allocated using MclassAllocResult(). The type of result buffer that you can specify depends on the copy operation (CopyType). Set this parameter to one of the following values.

function map For specifying the destination classification context
Click to summarizeValue Description
Click to summarize ContextClassifierCNNPredefinedId

Specifies the identifier of a predefined CNN classifier context (M_CLASSIFIER_CNN_PREDEFINED).

Click to summarize ContextClassifierTreeEnsembleId

Specifies the identifier of a tree ensemble classifier context (M_CLASSIFIER_TREE_ENSEMBLE).

Click to summarize ContextDatasetFeaturesId

Specifies the identifier of a features dataset context (M_DATASET_FEATURES).

Click to summarize ContextDatasetImagesId

Specifies the identifier of an images dataset context (M_DATASET_IMAGES).

DstIndex

Reserved for future expansion and must be set to M_DEFAULT.

CopyType

Specifies the type of copy operation to perform.

function map Specifies the type of copy operation
Click to summarizeValue Description
Click to summarize M_DEV_DATASET_PREDICT_SCORES

Specifies to copy the predicted scores of the development dataset established by MclassTrain(). To perform this copy operation, you must set the SrcResultClassId parameter to the identifier of a CNN or tree ensemble training result buffer and the DstContextClassId parameter to the identifier of a features or images dataset context.

If you used a tree ensemble classifier and did not specify a development dataset with MclassTrain(), nothing is copied.

(summarize)
Click to summarize M_OUT_OF_BAG_PREDICT_SCORES

Specifies to copy the predicted scores of the out-of-bag entries established by MclassTrain(). To perform this copy operation, you must set the SrcResultClassId parameter to the identifier of a tree ensemble training result buffer and the DstContextClassId parameter to the identifier of a features dataset context.

Note, bagging information is typically unreliable if your training dataset has augmented entries.

(summarize)
Click to summarize M_TRAIN_DATASET_PREDICT_SCORES

Specifies to copy the predicted scores of the training dataset established by MclassTrain(). To perform this copy operation, you must set the SrcResultClassId parameter to the identifier of a CNN or tree ensemble training result buffer and the DstContextClassId parameter to the identifier of a features or images dataset context.

(summarize)
Click to summarize M_TRAINED_CLASSIFIER_CNN

Specifies to copy the trained CNN classifier context established by MclassTrain(). To perform this copy operation, you must set the SrcResultClassId parameter to the identifier of a CNN training result buffer and the DstContextClassId parameter to the identifier of a CNN classifier context.

(summarize)
Click to summarize M_TRAINED_TREE_ENSEMBLE

Specifies to copy the trained tree ensemble classifier context established by MclassTrain(). To perform this copy operation, you must set the SrcResultClassId parameter to the identifier of a tree ensemble training result buffer and the DstContextClassId parameter to the identifier of a tree ensemble classifier context.

(summarize)
ControlFlag

Reserved for future expansion and must be set to M_DEFAULT.

Compilation information
Header Include mil.h.
Library Use mil.lib; milclass.lib.
DLL Requires mil.dll; milclass.dll.
DEV DATASET PREDICT SCORES OUT OF BAG PREDICT SCORES TRAIN DATASET PREDICT SCORES TRAINED CLASSIFIER CNN TRAINED TREE ENSEMBLE