could not load library libvx_tidl_rt.so

[niru-5]

Hi,

We are using docker environment suggestions from the dockerfile and also made sure we have python3.6 However, when we run the tfl python example, we get the following error ->

Running 3 Models - ['cl-tfl-mobilenet_v1_1.0_224', 'ss-tfl-deeplabv3_mnv2_ade20k_float', 'od-tfl-ssd_mobilenet_v2_300_float']

Running_Model :  cl-tfl-mobilenet_v1_1.0_224

Running_Model :  ss-tfl-deeplabv3_mnv2_ade20k_float

Running_Model :  od-tfl-ssd_mobilenet_v2_300_float
Process Process-3:
Process Process-1:
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tflite_runtime/interpreter.py", line 175, in load_delegate
    delegate = Delegate(library, options)
  File "/usr/local/lib/python3.6/dist-packages/tflite_runtime/interpreter.py", line 113, in __init__
    raise ValueError(capture.message)
ValueError: could not load library libvx_tidl_rt.so

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.6/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/usr/lib/python3.6/multiprocessing/process.py", line 93, in run
    self._target(*self._args, **self._kwargs)
  File "tflrt_delegate.py", line 174, in run_model
    experimental_delegates=[tflite.load_delegate(os.path.join(tidl_tools_path, 'tidl_model_import_tflite.so'), delegate_options)])
  File "/usr/local/lib/python3.6/dist-packages/tflite_runtime/interpreter.py", line 178, in load_delegate
    library, str(e)))
ValueError: Failed to load delegate from /resources/edgeai-tidl-tools/tidl_tools/tidl_model_import_tflite.so
could not load library libvx_tidl_rt.so
Process Process-2:
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tflite_runtime/interpreter.py", line 175, in load_delegate
    delegate = Delegate(library, options)
  File "/usr/local/lib/python3.6/dist-packages/tflite_runtime/interpreter.py", line 113, in __init__
    raise ValueError(capture.message)
ValueError: could not load library libvx_tidl_rt.so

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.6/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/usr/lib/python3.6/multiprocessing/process.py", line 93, in run
    self._target(*self._args, **self._kwargs)
  File "tflrt_delegate.py", line 174, in run_model
    experimental_delegates=[tflite.load_delegate(os.path.join(tidl_tools_path, 'tidl_model_import_tflite.so'), delegate_options)])
  File "/usr/local/lib/python3.6/dist-packages/tflite_runtime/interpreter.py", line 178, in load_delegate
    library, str(e)))
ValueError: Failed to load delegate from /resources/edgeai-tidl-tools/tidl_tools/tidl_model_import_tflite.so
could not load library libvx_tidl_rt.so
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tflite_runtime/interpreter.py", line 175, in load_delegate
    delegate = Delegate(library, options)
  File "/usr/local/lib/python3.6/dist-packages/tflite_runtime/interpreter.py", line 113, in __init__
    raise ValueError(capture.message)
ValueError: could not load library libvx_tidl_rt.so

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.6/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/usr/lib/python3.6/multiprocessing/process.py", line 93, in run
    self._target(*self._args, **self._kwargs)
  File "tflrt_delegate.py", line 174, in run_model
    experimental_delegates=[tflite.load_delegate(os.path.join(tidl_tools_path, 'tidl_model_import_tflite.so'), delegate_options)])
  File "/usr/local/lib/python3.6/dist-packages/tflite_runtime/interpreter.py", line 178, in load_delegate
    library, str(e)))
ValueError: Failed to load delegate from /resources/edgeai-tidl-tools/tidl_tools/tidl_model_import_tflite.so
could not load library libvx_tidl_rt.so
^CTraceback (most recent call last):
  File "tflrt_delegate.py", line 257, in <module>
    nthreads = join_one(nthreads)
  File "tflrt_delegate.py", line 239, in join_one
    sem.acquire()

However, we have checked that the TIDL_TOOLS_PATH is set and has the necessary libs.

ls ${TIDL_TOOLS_PATH}
device_config.cfg        libvx_tidl_rt.so         tidl_graphVisualiser.out           tidl_model_import.so
itidl_rt.h               libvx_tidl_rt.so.1.0     tidl_graphVisualiser_runtimes.out  tidl_model_import_tflite.so
itvm_rt.h                osrt_deps                tidl_model_import_onnx.so          yaml-cpp
libtidl_onnxrt_EP.so     PC_dsp_test_dl_algo.out  tidl_model_import.out
libtidl_tfl_delegate.so  ti_cnnperfsim.out        tidl_model_import_relay.so

Is there anything more we can do to debug the issue?

Thank You Niranjan

[GesilaA]

You may need to run commands below

export DEVICE=j7
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$TIDL_TOOLS_PATH:$TIDL_TOOLS_PATH/osrt_deps

[niru-5]

Hi,

That did solve the problem.

Thanks, Niranjan