TypeError: Input 'shape' of 'ScatterNd' Op has type int64 that does not match type int32 of argument 'indices'.

Trying to test Local Implicit sample I've found this issue:

Performing latent grid optimization...
Traceback (most recent call last):
  File "C:\Users\Victor\anaconda3\envs\LocalImplicit\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 508, in _apply_op_helper
    values = ops.convert_to_tensor(
  File "C:\Users\Victor\anaconda3\envs\LocalImplicit\lib\site-packages\tensorflow\python\profiler\trace.py", line 183, in wrapped
    return func(*args, **kwargs)
  File "C:\Users\Victor\anaconda3\envs\LocalImplicit\lib\site-packages\tensorflow\python\framework\ops.py", line 1662, in convert_to_tensor
    raise ValueError(
ValueError: Tensor conversion requested dtype int32 for Tensor with dtype int64: <tf.Tensor 'Const_3:0' shape=(2,) dtype=int64>

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "D:\Dependencias\LocalImplicitGrid\graphics\reconstruct_geometry.py", line 108, in <module>
    app.run(main)
  File "C:\Users\Victor\anaconda3\envs\LocalImplicit\lib\site-packages\absl\app.py", line 312, in run
    _run_main(main, args)
  File "C:\Users\Victor\anaconda3\envs\LocalImplicit\lib\site-packages\absl\app.py", line 258, in _run_main
    sys.exit(main(argv))
  File "D:\Dependencias\LocalImplicitGrid\graphics\reconstruct_geometry.py", line 88, in main
    v, f, _, _ = rec.encode_decoder_one_scene(
  File "D:\Dependencias\LocalImplicitGrid\graphics\tensorflow_graphics\projects\local_implicit_grid\core\reconstruction.py", line 351, in encode_decoder_one_scene
    goptim = LIGOptimizer(
  File "D:\Dependencias\LocalImplicitGrid\graphics\tensorflow_graphics\projects\local_implicit_grid\core\reconstruction.py", line 61, in __init__
    self._init_graph()
  File "D:\Dependencias\LocalImplicitGrid\graphics\tensorflow_graphics\projects\local_implicit_grid\core\reconstruction.py", line 106, in _init_graph
    self.feat_grid = tf.scatter_nd(self.occ_idx_flat_,
  File "C:\Users\Victor\anaconda3\envs\LocalImplicit\lib\site-packages\tensorflow\python\ops\gen_array_ops.py", line 9185, in scatter_nd
    _, _, _op, _outputs = _op_def_library._apply_op_helper(
  File "C:\Users\Victor\anaconda3\envs\LocalImplicit\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 546, in _apply_op_helper
    raise TypeError(
TypeError: Input 'shape' of 'ScatterNd' Op has type int64 that does not match type int32 of argument 'indices'.

Any tip on this?

Thank you in advance.

Start with graphics/reconstruct_geometry.py and follow the Local Implicit sample into core/reconstruction.py, especially LIGOptimizer._init_graph at line 106 where tf.scatter_nd fails. Reproduce the sample's latent grid optimization and inspect the dtypes of the indices and shape inputs. Done means reconstruction proceeds past this ScatterNd error.