write_dl_modelT_write_dl_modelWriteDlModelWriteDlModelwrite_dl_model (算子)
名称
write_dl_modelT_write_dl_modelWriteDlModelWriteDlModelwrite_dl_model — 在文件中写入一个深度学习模型。
签名
描述
write_dl_modelwrite_dl_modelWriteDlModelWriteDlModelWriteDlModelwrite_dl_model writes the deep learning model
DLModelHandleDLModelHandleDLModelHandleDLModelHandleDLModelHandledlmodel_handle to the file given by FileNameFileNameFileNameFileNamefileNamefile_name.
Please note that the values of the runtime specific parameters 'gpu'"gpu""gpu""gpu""gpu""gpu",
'runtime'"runtime""runtime""runtime""runtime""runtime", and 'runtime_init'"runtime_init""runtime_init""runtime_init""runtime_init""runtime_init" are not written.
The default HALCON file extension for deep learning models is
'.hdl'.
The model can be read with read_dl_modelread_dl_modelReadDlModelReadDlModelReadDlModelread_dl_model.
For further explanations to deep learning models in HALCON,
see the chapter Deep Learning / Model.
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。
参数
DLModelHandleDLModelHandleDLModelHandleDLModelHandleDLModelHandledlmodel_handle (输入控制) dl_model → HDlModel, HTupleHHandleHTupleHtuple (handle) (IntPtr) (HHandle) (handle)
Handle of the deep learning model.
FileNameFileNameFileNameFileNamefileNamefile_name (输入控制) filename.write → HTuplestrHTupleHtuple (string) (string) (HString) (char*)
Filename
File extension:
.hdl
结果
If the parameters are valid, the operator write_dl_modelwrite_dl_modelWriteDlModelWriteDlModelWriteDlModelwrite_dl_model
returns the value 2 (
H_MSG_TRUE)
. If necessary, an exception is raised.
可能的前趋
read_dl_modelread_dl_modelReadDlModelReadDlModelReadDlModelread_dl_model,
train_dl_model_batchtrain_dl_model_batchTrainDlModelBatchTrainDlModelBatchTrainDlModelBatchtrain_dl_model_batch,
train_dl_model_anomaly_datasettrain_dl_model_anomaly_datasetTrainDlModelAnomalyDatasetTrainDlModelAnomalyDatasetTrainDlModelAnomalyDatasettrain_dl_model_anomaly_dataset,
set_dl_model_paramset_dl_model_paramSetDlModelParamSetDlModelParamSetDlModelParamset_dl_model_param
模块
基础。该算子采用动态许可机制(详见《安装指南》)。所需模块取决于算子的具体使用场景:
3D计量学、光学字符识别/光学字符验证、匹配、深度学习推理