Point Transformer training for SemanticKitti dataset

### Checklist

- [X] I have searched for [similar issues](https://github.com/isl-org/Open3D-ML/issues).
- [X] I have tested with the [latest development wheel](http://www.open3d.org/docs/latest/getting_started.html#development-version-pip).
- [X] I have checked the [release documentation](http://www.open3d.org/docs/release/) and the [latest documentation](http://www.open3d.org/docs/latest/) (for `main` branch).

### My Question

> import open3d.ml as ml3d
> from open3d.ml.torch.models import RandLANet, PointTransformer
> from open3d.ml.torch.pipelines import SemanticSegmentation
>
> dataset = ml3d.datasets.SemanticKITTI(dataset_path='/home/jmgoo/projects/DATA/SemanticKitti', use_cache=True)
> model = PointTransformer()
>
> pipeline = SemanticSegmentation(model=model, dataset=dataset, **cfg.pipeline)
> t0 = time.time()
> pipeline.run_train()
> print("Training time: ", time.time()-t0)

I run the above code, but I faced "AttributeError: 'ConfigDict' object has no attribute 'optimizer'". I believe there is no yaml file yet.

How do I solve this problem?