Also team! I am facing a weird error and it started all of a sudden, nothing changed from code wise TypeVarTuple.__init__() got an unexpected keyword argument 'default'. Dataset 'RAW_CAMPUS' must only contain arguments valid for the constructor of 'kedro.extras.datasets.pandas.csv_dataset.CSVDataSet'. Below is the Catalog entry RAW_CAMPUS: type: pandas.CSVDataSet filepath: s3://464-shapeeda/raw/campus/Campus.csv Any help is much appreciated!!

can you please confirm the kedro-datasets version ?

Hi Ravi! Sure Name: kedro-datasets Version: 2.1.0

I am not sure how this was working before but can you refer to - https://docs.kedro.org/projects/kedro-datasets/en/kedro-datasets-2.1.0.post1/api/kedro_datasets.pandas.CSVDataset.html You need to use CSVData*s*et instead of CSVData*S*et. You can also find YAML API there. Can you try updating the usage and see if it gets fixed ? Also, which kedro version are you using ?

With the same versions if I run locally this is running fine Ravi, but let us give it a try

Hey Pradeep, which version of Kedro are you on? The recent versions of Kedro do not have

kedro.extras.datasets

but the datasets come from

kedro-datasets

hello @Ankita Katiyar we are using Kedro 0.18.11, locally it runs fine

Where is this error occurring? I suspect it has an old version of Kedro

Hi Ankita! It's occurring at the start of the pipeline, I guess while loading the catalog, it's not even loading datasets. And the catalog has other entries and the error is happening for a dataset that irrelevant to the specific pipeline

Could you check with “kedro —version” or “kedro info” command on the platform you’re facing this issue?

This is the Kedro version we have in the requirements.txt that is being used to create the .whl file kedro==0.18.11

i think then the issue might be that you need to update the catalog.yml to “pandas.CSVDataset” with a lowecase “s” like Ravi mentioned. Because it’s “DataSet” in your catalog, it’s pulling the CSV dataset from extras.datasets which are old and not updated instead of from kedro-datasets package

Cool, thanks Ankita! Will try and keep posted here