* @return 0 if ok. < 0 if error.
*/
- int (*train) (const GstTensorTrainerFramework * self,
+ int (*start) (const GstTensorTrainerFramework * self,
const GstTensorTrainerProperties * prop, void *private_data);
- /**< tensor_trainer call this to train the model
+ /**< tensor_trainer call this to start training the model
* @param[in] prop read-only property values
* @param[in] private_data, a subplugin may save its internal private data here.
* @return 0 if ok. < 0 if error.
*/
- int (*invoke) (const GstTensorTrainerFramework * self,
+ int (*push_data) (const GstTensorTrainerFramework * self,
const GstTensorTrainerProperties * prop,
void *private_data, const GstTensorMemory * input);
- /**< tensor_trainer call this to send tensor data to subplugin, subplugin constructs a data set using input.
+ /**< tensor_trainer call this to push tensor data to subplugin, subplugin constructs a data set using input.
* @param[in] prop read-only property values
* @param[in] private_data, a subplugin may save its internal private data here.
* @param[in] input The array of input tensors. Allocated and filled by tensor_trainer