Clean Missing data - Not able to select by names

Srinivasa Rama Subrahmanyam Jonnalagadda 5

Hi,

Below is my pipeline and i am not able to select columns by name for "Clean missing data" transformation. User's image

User's image

YutongTie-MSFT 47,416 Reputation points

2023-07-25T00:51:00.18+00:00

Hello @Srinivasa Rama Subrahmanyam Jonnalagadda

Thanks for reaching out to us.

The auto-suggestion for column names in Azure Machine Learning only works when the upstream node is a data node and schema exists for the data. This is because the column names are inferred from the schema of the data node.

If the upstream node is not a data node or if the schema is not available, you may need to manually enter the column names or use the column selector to choose the columns you want to use. Additionally, you can use the "Edit Metadata" component to specify the metadata for the columns, such as whether they are categorical or numeric, which can help downstream components to better understand the data.

Please check on your upstream node if you want to use it.

I hope this helps.

Regards,

Yutong
Srinivasa Rama Subrahmanyam Jonnalagadda 5 Reputation points

2023-07-25T03:59:12.5166667+00:00

Hi @YutongTie-MSFT thanks for the quick reply. The confusion i have is the upstream node "select columns in dataset" should pass on selected columns down stream at least, if not entire dataset. DO i need to change any settings in "select columns in dataset" to pass selected columns downstream - selectable.
Srinivasa Rama Subrahmanyam Jonnalagadda 5 Reputation points

2023-07-25T04:18:12.32+00:00

Hi @YutongTie-MSFT thanks for quick reply. Upstream i have "select columns from dataset", and i assume this node (Not sure whether its a data node or not), should pass down selected columns downstream in such a way that they are selectable. Or how do i change settings of "select columns from dataset" to make it happen.
YutongTie-MSFT 47,416 Reputation points

2023-08-01T18:19:51.8633333+00:00

Hello @Srinivasa Rama Subrahmanyam Jonnalagadda

You can have a try, at the meantime, could you please share your input columns format so that we can look into it to see how we can make it happen? Thanks.
Brian Van Steen 20 Reputation points

2024-05-14T14:26:39.42+00:00

@YutongTie-MSFT
I have the same issue.
I am using the Sample Microsoft data; Restaurant Ratings and Restaurant Feature Data.
I am following a basic tutorial, to join the data from the two data sets. For selecting the columns, these were auto-populated.
But for joining the data, I received the same message, and had to manually type in 'placeID'.
I actually ran the model at the Join Data stage, and it was successful.
But now for the Clean Missing Data, I again get the same message "Auto-suggestion for column names....".
Is there some bug in the Designer tool? These are simple .csv files using the drag-and-drop components, so I should not have to manually type column names.

Share via

Clean Missing data - Not able to select by names