TypeError: 'type' object is not subscriptable

[Warvito]

After adding the notations, started to get this error

Traceback (most recent call last):
  File "/project/src/python/training/train_aekl.py", line 8, in <module>
    from generative.networks.nets import AutoencoderKL
  File "/usr/local/lib/python3.8/dist-packages/generative/networks/nets/__init__.py", line 18, in <module>
    from .vqvae import VQVAE
  File "/usr/local/lib/python3.8/dist-packages/generative/networks/nets/vqvae.py", line 22, in <module>
    from generative.networks.layers.vector_quantizer import EMAQuantizer, VectorQuantizer
  File "/usr/local/lib/python3.8/dist-packages/generative/networks/layers/__init__.py", line 12, in <module>
    from .vector_quantizer import EMAQuantizer, VectorQuantizer
  File "/usr/local/lib/python3.8/dist-packages/generative/networks/layers/vector_quantizer.py", line 20, in <module>
    class EMAQuantizer(nn.Module):
  File "/usr/local/lib/python3.8/dist-packages/generative/networks/layers/vector_quantizer.py", line 87, in EMAQuantizer
    def quantize(self, inputs: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
TypeError: 'type' object is not subscriptable

[ericspod]

This is an issue with Python 3.8. So long as we support that version we have to use the typing annotations I think.