trainf_ocr_class_boxT_trainf_ocr_class_boxTrainfOcrClassBoxTrainfOcrClassBoxtrainf_ocr_class_box (算子)
名称
trainf_ocr_class_boxT_trainf_ocr_class_boxTrainfOcrClassBoxTrainfOcrClassBoxtrainf_ocr_class_box — 在训练文件的帮助下训练 OCR 分类器。
警告
trainf_ocr_class_boxtrainf_ocr_class_boxTrainfOcrClassBoxTrainfOcrClassBoxTrainfOcrClassBoxtrainf_ocr_class_box 已过时，仅出于向后兼容性考虑而保留。 New applications should use the
MLP, SVM or CNN based operators instead.
签名
描述
算子 trainf_ocr_class_boxtrainf_ocr_class_boxTrainfOcrClassBoxTrainfOcrClassBoxTrainfOcrClassBoxtrainf_ocr_class_box trains the classifier
OcrHandleOcrHandleOcrHandleOcrHandleocrHandleocr_handle via the indicated training files.
Any number of files can be indicated. The parameter AvgConfidenceAvgConfidenceAvgConfidenceAvgConfidenceavgConfidenceavg_confidence
provides information about the success of the training: It contains
the average confidence of the trained characters measured by a
re-classification. The confidence of mismatched characters
is set to 0 (thus, the average confidence will be decreased significantly).
Please, note that training characters that have no corresponding class in the
classifier OcrHandleOcrHandleOcrHandleOcrHandleocrHandleocr_handle are discarded.
注意
The names of the characters in the file must fit the
network.
执行信息
- 多线程类型：可重入（与非独占算子并行运行）。
- 多线程作用域：全局（可从任何线程调用）。
- 未采用并行化处理。
此算子修改后续输入参数的状态：
在执行此算子时，若该参数值需在多个线程间使用，则必须对其访问进行同步。
参数
OcrHandleOcrHandleOcrHandleOcrHandleocrHandleocr_handle (输入控制，状态被修改) ocr_box → HOCRBox, HTupleHHandleHTupleHtuple (handle) (IntPtr) (HHandle) (handle)
ID of the desired OCR-network.
TrainingFileTrainingFileTrainingFileTrainingFiletrainingFiletraining_file (输入控制) filename.read(-array) → HTupleMaybeSequence[str]HTupleHtuple (string) (string) (HString) (char*)
Names of the training files.
默认值：
'train_ocr'
"train_ocr"
"train_ocr"
"train_ocr"
"train_ocr"
"train_ocr"
File extension:
.trf, .otr
AvgConfidenceAvgConfidenceAvgConfidenceAvgConfidenceavgConfidenceavg_confidence (输出控制) real → HTuplefloatHTupleHtuple (real) (double) (double) (double)
Average confidence during a re-classification
of the trained characters.
示例（C）
HTuple TrainingFile, OcrHandle, AvgConfidence;
T_create_ocr_class_box(WidthPattern,HeightPattern,Interpolation,
Features,\Character,&OcrHandle);
create_tuple(&TrainingFile,2);
set_s(TrainingFile,"data1",0);
set_s(TrainingFile,"data2",1);
T_trainf_ocr_class_box(OcrHandle,TrainingFile,&AvgConfidence);
结果
If the file name is correct and the data fit the network,
the operator trainf_ocr_class_boxtrainf_ocr_class_boxTrainfOcrClassBoxTrainfOcrClassBoxTrainfOcrClassBoxtrainf_ocr_class_box 返回值 2 ( H_MSG_TRUE )。否则将抛出异常。
可能的前趋
create_ocr_class_boxcreate_ocr_class_boxCreateOcrClassBoxCreateOcrClassBoxCreateOcrClassBoxcreate_ocr_class_box,
read_ocrread_ocrReadOcrReadOcrReadOcrread_ocr
可能的后继
traind_ocr_class_boxtraind_ocr_class_boxTraindOcrClassBoxTraindOcrClassBoxTraindOcrClassBoxtraind_ocr_class_box,
write_ocrwrite_ocrWriteOcrWriteOcrWriteOcrwrite_ocr,
do_ocr_multido_ocr_multiDoOcrMultiDoOcrMultiDoOcrMultido_ocr_multi,
do_ocr_singledo_ocr_singleDoOcrSingleDoOcrSingleDoOcrSingledo_ocr_single
替代
traind_ocr_class_boxtraind_ocr_class_boxTraindOcrClassBoxTraindOcrClassBoxTraindOcrClassBoxtraind_ocr_class_box
模块
光学字符识别/光学字符验证