I have read tflm_cifar10 in SDK. In general, we need to convert Tflite model into C array. Now I want to evaluate TFLM model (C array, not .tflite model) in the PC to see whether accuracy is dropped. The reason I want to run in PC to save time. Do you have any source code for classification model such as for Cifar10?
I move from Python to C/C++ and something is not familiar. Thank you.
Hi @nnxxpp ,
Are you looking for some method like below?
https://www.tensorflow.org/lite/microcontrollers/build_convert#convert_to_a_c_array
Hope that helps,
Have a great day,
Kan
-------------------------------------------------------------------------------
Note:
- If this post answers your question, please click the "Mark Correct" button. Thank you!
- We are following threads for 7 weeks after the last post, later replies are ignored
Please open a new thread and refer to the closed one, if you have a related question at a later point in time.
-------------------------------------------------------------------------------
Thanks. I have already converted tflite to tflite .C array. I need to guide to evaluate .tflite .C array on PC. In general, we directly deploy .tflite C array on board without performance validation. Because I could not find any guide to do this, so I opened this topic.