kedro.extras.datasets.pandas.ExcelDataSet

ExcelDataSet loads/saves data from/to a Excel file using an underlying filesystem (e.g.: local, S3, GCS). It uses pandas to handle the Excel file.

Example usage for the YAML API:

rockets:
  type: pandas.ExcelDataSet
  filepath: gcs://your_bucket/rockets.xlsx
  fs_args:
    project: my-project
  credentials: my_gcp_credentials
  save_args:
    sheet_name: Sheet1
  load_args:
    sheet_name: Sheet1

shuttles:
  type: pandas.ExcelDataSet
  filepath: data/01_raw/shuttles.xlsx

Example usage for the Python API:

>>> from kedro.extras.datasets.pandas import ExcelDataSet
>>> import pandas as pd
>>>
>>> data = pd.DataFrame({'col1': [1, 2], 'col2': [4, 5],
>>>                      'col3': [5, 6]})
>>>
>>> data_set = ExcelDataSet(filepath="test.xlsx")
>>> data_set.save(data)
>>> reloaded = data_set.load()
>>> assert data.equals(reloaded)

To save a multi-sheet Excel file, no special save_args are required. Instead, return a dictionary of Dict[str, pd.DataFrame] where the string keys are your sheet names.

Example usage for the YAML API for a multi-sheet Excel file:

trains:
  type: pandas.ExcelDataSet
  filepath: data/02_intermediate/company/trains.xlsx
  load_args:
    sheet_name: [Sheet1, Sheet2, Sheet3]

Example usage for the Python API for a multi-sheet Excel file:

>>> from kedro.extras.datasets.pandas import ExcelDataSet
>>> import pandas as pd
>>>
>>> dataframe = pd.DataFrame({'col1': [1, 2], 'col2': [4, 5],
>>>                      'col3': [5, 6]})
>>> another_dataframe = pd.DataFrame({"x": [10, 20], "y": ["hello", "world"]})
>>> multiframe = {"Sheet1": dataframe, "Sheet2": another_dataframe}
>>> data_set = ExcelDataSet(filepath="test.xlsx", load_args = {"sheet_name": None})
>>> data_set.save(multiframe)
>>> reloaded = data_set.load()
>>> assert multiframe["Sheet1"].equals(reloaded["Sheet1"])
>>> assert multiframe["Sheet2"].equals(reloaded["Sheet2"])