MyungJoo Ham

> HI, > > * I think this project is very useful for myself. I want to add the supporte for rockchip(rk3399 pro) , huawei( atlas 200dk) and Xilinx. I...

> I am also want to integrate [openncc](www.openncc.com) to NNStreamer, could you show me more how to write a subplugin? as in https://github.com/nnstreamer/nnstreamer/issues/3486#issuecomment-915163227 , a tensor-filter subplugin code would be...

with no further follow-ups, I assume this is resolved. Please reopen if there are things to be discussed further.

Check if your TVM .so works with GPU without nnstreamer, first.

Your python code gives inputs as 512:512:3, float32. Your nnstreamer pipeline gives inputs as 128:128:3, float32. It appears that you are not testing with the same model file and you...

Will you try to update ``` t_raw. ! queue leaky=2 max-size-buffers=2 ! videoscale ! video/x-raw,width=512,height=512! ``` ==> ``` t_raw. ! queue leaky=2 max-size-buffers=2 ! videoscale ! video/x-raw,format=RGB,width=512,height=512! ``` It appears...

Draft 220602_MJ_A (WIP) Filter Name: ```tensor_trainer``` Subplugin Required: true - mode = separated streams - SINKPAD (input) 0: X (input tensors) - SINKPAD (input) 1: y (supposed output) - mode...

Please write proposals of drafts as usage cases of pipeline topology (component view). We may need to include a repository/database element for training data. (e.g., GST-Pravega w/ other/tensors supprot)

Milestone - Design iteration: 2023.1. - Working-PoC: 2023-1Q - Release with basic features: 7.5 M1 - nntrainer compatibility - C-API - example pipelines & apps - Release with full features:...

If you are going to increase it, let's try 8. :) We may let each tensor\_* element declare its supported RANK. Most elements will be trivial for this change; however,...