Specification of dataset type in training docs (`06_jupyter_notebook_workflow.md`) as `pandas.JSONDataSet` throws an `AttributeError` when saving

[abhi8893]

Description

The 06_jupyter_notebook_workflow.md in the training_docs has the following section:

Put the following dataset entry in conf/base/catalog.yml:

my_dataset:
  type: pandas.JSONDataSet
  filepath: data/01_raw/my_dataset.json

Next, you need to reload Kedro variables by calling %reload_kedro line magic in your Jupyter notebook.

Finally, you can save the data by executing the following command:

my_dict = {"key1": "some_value", "key2": None}
catalog.save("my_dataset", my_dict)

Error

The specified type of the dataset i.e. type: pandas.JSONDataSet seems throw an AttributeError when saving the data:

Traceback:

2021-11-10 23:11:44,722 - kedro.io.data_catalog - INFO - Saving data to `my_dataset` (JSONDataSet)...
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
/Volumes/GoogleDrive/My Drive/projects/kedro-training/venv/lib/python3.8/site-packages/kedro/io/core.py in save(self, data)
    237             self._logger.debug("Saving %s", str(self))
--> 238             self._save(data)
    239         except DataSetError:

/Volumes/GoogleDrive/My Drive/projects/kedro-training/venv/lib/python3.8/site-packages/kedro/extras/datasets/pandas/json_dataset.py in _save(self, data)
    161         with self._fs.open(save_path, **self._fs_open_args_save) as fs_file:
--> 162             data.to_json(path_or_buf=fs_file, **self._save_args)
    163 

AttributeError: 'dict' object has no attribute 'to_json'

The above exception was the direct cause of the following exception:

DataSetError                              Traceback (most recent call last)
/var/folders/ps/m62g53713k7_knw76b1s566r0000gn/T/ipykernel_55726/2517118377.py in <module>
      1 my_dict = {"key1": "some_value", "key2": None}
----> 2 catalog.save("my_dataset", my_dict)

/Volumes/GoogleDrive/My Drive/projects/kedro-training/venv/lib/python3.8/site-packages/kedro/io/data_catalog.py in save(self, name, data)
    447 
    448         func = self._get_transformed_dataset_function(name, "save", dataset)
--> 449         func(data)
    450 
    451         version = (

/Volumes/GoogleDrive/My Drive/projects/kedro-training/venv/lib/python3.8/site-packages/kedro/io/core.py in save(self, data)
    623         save_version = self.resolve_save_version()  # Make sure last save version is set
    624         try:
--> 625             super().save(data)
    626         except (FileNotFoundError, NotADirectoryError) as err:
    627             # FileNotFoundError raised in Win, NotADirectoryError raised in Unix

/Volumes/GoogleDrive/My Drive/projects/kedro-training/venv/lib/python3.8/site-packages/kedro/io/core.py in save(self, data)
    243         except Exception as exc:
    244             message = f"Failed while saving data to data set {str(self)}.\n{str(exc)}"
--> 245             raise DataSetError(message) from exc
    246 
    247     def __str__(self):

DataSetError: Failed while saving data to data set JSONDataSet(filepath=/Volumes/GoogleDrive/My Drive/projects/kedro-training/spaceflights/data/01_raw/my_dataset.json, load_args={}, protocol=file, save_args={}).
'dict' object has no attribute 'to_json'

Possible Resolution

Specifying the dataset type as type: json.JSONDataSet resolves the problem.