[dataset] The dataset folder of zipped dataset is added twice

[fcdl94]

Having a dataset zipped with the following structure and selecting dataset_dirs = datasets

dataset_name/
  train/
  valid/
  etc.

the resulting dataset will be put in datasets/datasets/zip_name/dataset_name.
The zip_name/dataset_name is ok, but the double datasets is weird.

The code that generates the error is here:

  else:
      dataset_name = dataset_name.split(".")[0]
      _dest_path = os.path.join(self.datasets_dir, dataset_name)
      dataset_path = extract_archive(dataset_path, _dest_path)
      logger.info(f"Extracted archive: {dataset_path}, {os.listdir(dataset_path)}")

[CuriousDolphin]

@fcdl94 this is resolved by #161 ?