rasterizer_op.so: undefined symbol

Error when running inverse_rendering.ipynb on colab with TF version 2.15.0 (default):

https://colab.research.google.com/github/tensorflow/graphics/blob/master/tensorflow_graphics/notebooks/inverse_rendering.ipynb

---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/tensorflow_graphics/rendering/kernels/rasterization_backend.py](https://localhost:8080/#) in <module>
     32 try:
---> 33   from tensorflow_graphics.rendering.kernels import gen_rasterizer_op as render_ops
     34 except ImportError:

ImportError: cannot import name 'gen_rasterizer_op' from 'tensorflow_graphics.rendering.kernels' (/usr/local/lib/python3.10/dist-packages/tensorflow_graphics/rendering/kernels/__init__.py)

During handling of the above exception, another exception occurred:

NotFoundError                             Traceback (most recent call last)
13 frames
[/usr/local/lib/python3.10/dist-packages/tensorflow/python/framework/load_library.py](https://localhost:8080/#) in load_op_library(library_filename)
     52     RuntimeError: when unable to load the library or get the python wrappers.
     53   """
---> 54   lib_handle = py_tf.TF_LoadLibrary(library_filename)
     55   try:
     56     wrappers = _pywrap_python_op_gen.GetPythonWrappers(

NotFoundError: /usr/local/lib/python3.10/dist-packages/tensorflow_graphics/rendering/kernels/rasterizer_op.so: undefined symbol: _ZNK10tensorflow8OpKernel11TraceStringERKNS_15OpKernelContextEb

!pip freeze | grep tensorflow

tensorflow-addons==0.23.0
tensorflow-datasets==4.9.4
tensorflow-estimator==2.15.0
tensorflow-gcs-config==2.15.0
tensorflow-graphics==2021.12.3
tensorflow-hub==0.16.1
tensorflow-io-gcs-filesystem==0.36.0
tensorflow-metadata==1.14.0
tensorflow-probability==0.23.0

Downgrading TF to 2.8 allows it to run:

!pip uninstall tensorflow -y
!pip install tensorflow~=2.8.0
!pip install tensorflow_graphics

Start by running tensorflow_graphics/notebooks/inverse_rendering.ipynb with TensorFlow 2.15.0 and inspect tensorflow_graphics/rendering/kernels/rasterization_backend.py, which imports the generated rasterizer operation. Check how rasterizer_op.so is built and loaded, using the reported undefined symbol as the failure condition. Done means the notebook imports and runs with the listed TensorFlow version without requiring a downgrade.