Hemanth-Jonnala

@abucka can you please share the .exe files in the out directory. Thanks!

Hello @baghbidi can you please share the posenet_shared_lib.dll file here. Thanks!

edgetpu rutime 14 does not work on windows. So can you please try with release-frogfish and share the build instructions here: https://github.com/google-coral/libcoral/tags Thanks!

Have you compiled it on Windows Machine or MacOS? Can you please share the build instruction steps to reproduce the issue.. Thanks!

I have tried to build posenet decoder on Windows similar to what we did for [MacOS](https://github.com/google-coral/project-bodypix/issues/22#issuecomment-940221257). Unfortunately, I am not able to [build](https://github.com/hjonnala/snippets/blob/main/posenet/windows/log1.txt) it successfully due to incomplete build setup...

Hi @Snail124 this project hasn't been implemented for Dev Board Mini. > ```shell > RuntimeError: Failed to prepare for TPU. Failed precondition: Package requires runtime version (14), which is newer...

please try flashing the board using force boot method: https://coral.ai/docs/dev-board-mini/reflash/#force-boot-into-fastboot-mode

Hello @HanChangHun It could be due to input output latency. can you please share the latency results with [single model benchmark](https://github.com/google-coral/libcoral/blob/master/coral/single_model_benchmark.cc) for each segment file in txt file. Thanks! https://github.com/google-coral/edgetpu/issues/593#issuecomment-1137929277

Hmm..profile-based partitioner is not intended to perform the partitioning in a single edge tpu. Please check this page for details and requirements to use this tool: https://github.com/google-coral/libcoral/blob/master/coral/tools/partitioner/README.md. Thanks!

Can you please try [this code](https://github.com/hjonnala/snippets/blob/main/profiling/profiling_based_partitioner.cc) with two TPUs and with two segments on [inception v3 model](https://raw.githubusercontent.com/google-coral/test_data/master/inception_v3_299_quant.tflite) and share the logs and [single model benchmark](https://github.com/google-coral/libcoral/blob/master/coral/single_model_benchmark.cc) results for output models. Thanks!