Jihoon Lee

@kparichay could you check if the issue has been resolved?

### Background This was added to mitigate slight semantic difference between nnstreamer in gstreamer behavior and ml_api_singleshot behavior. [Long story](https://github.com/nnstreamer/api/pull/105/commits/efbfae9124be497df5b715e390b94a8a3ea90917) short, nntrainer inference filter [wants to](https://github.com/nnstreamer/nntrainer/pull/1711#discussion_r755185603) use default dimension if...

opinion about option 2: Policy B is direct interpretation of `createOrExtend()`, i think the term `link` lacks of the create sections. Advocates about unique / shared (at option 1, 3):...

resnet 18 test (number of params 11,230,948) unit: microseconds 10 iteration. ### batch size: 10 ``` key avg min max sum ============================================================================================================== forward + backward 1084403 1031972 1185450 10844036 offload...

No, it was not done is parallel. roughly ``` forward() && backward() saveModel() // with saving inputs as well readModel() // with reading inputs as well ``` So there will...

additional list - fc check if unit is empty at finalize, and issue proper warning

> AFAIK, we will decide when ths tensors are off-loaded with tensor life span and execution order. I think using lifespan outside of tensor is not possible because we simply...

> Consider two tensors, which have the same execution order usage. > > * Tensor 1 : exec_order - x1, x2 (x2 > x1) > * Tensor 2 : exec_order...

I don't like it but as a quick fix, we can have [layer 1] shared_from=layer_1 [layer 2] shared_from=layer_1 [layer 3] shared_from=layer_1

- update bn channel wise testing https://github.com/nnstreamer/nntrainer/pull/1584#discussion_r716486794 - Test for inplace mode https://github.com/nnstreamer/nntrainer/pull/1587#discussion_r716311440