[WIP] PyTorch-based training on Gaudi HPU

[jmduarte]

• PyTorch-based training on Gaudi HPU Issue when importing torch_cluster

>>> import torch_cluster
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python3.10/dist-packages/torch_cluster/__init__.py", line 18, in <module>
    torch.ops.load_library(spec.origin)
  File "/usr/local/lib/python3.10/dist-packages/torch/_ops.py", line 852, in load_library
    ctypes.CDLL(path)
  File "/usr/lib/python3.10/ctypes/__init__.py", line 374, in __init__
    self._handle = _dlopen(self._name, mode)
OSError: /usr/local/lib/python3.10/dist-packages/torch_cluster/_version_cpu.so: undefined symbol: _ZN5torch3jit17parseSchemaOrNameERKSs

[jpata]

torch_cluster comes in via torch_geometric.

We could get rid of the torch_geometric dependency by removing the GravNet code and always just doing 3D padding of the model: https://github.com/jpata/particleflow/blob/main/mlpf/pyg/PFDataset.py#L131. Technically we don't need it, it's a leftover.

[jpata]

FYI in this PR I removed the torch_geometric dependency: https://github.com/jpata/particleflow/pull/310

[jpata]

https://github.com/jpata/particleflow/pull/310 is merged, you can try updating.