Which will be the minor version for Skill creation?

If we trained on the base model what will be the minor version of the package to be selected 0 or 1(trained)?

Hi @Divya_Salve

You can use .1 version. When a training pipeline or a full pipeline executes successfully on a package version, a new minor version is created. For example, if I have uploaded a package (version 1.0), and start a training pipeline, version 1.1 is displayed after completion in the ML Package Details page as below:

Also, you can refer to the following document:

Hope this helps.
Regards.

1 Like

okay Thank you so we need to select the minor version as .1 as thats the trained one.
Also , if we run a new training pipeline with some new data, will the minor version be 0 or this .1?

@Divya_Salve

Yes. The minor versions keep on updating in an incremental fashion whenever you train, starting from .1

Hope this helps.
Best Regards.

1 Like

yes but we need to select 0 as minor version when creating a skill right?

@Divya_Salve

Initially, you will be selecting 0. If any pipeline runs, you can move up in the versions. You’ll also be able to roll back to the previous versions as per the requirements.

Best Regards.

2 Likes

@Divya_Salve

So here is what you need to do

select the latest version if you want to retrain the already trained model with new data

if you want to train the model fresh with new data leaving the old then select 0

hope this helps

cheers

2 Likes

Can we loop through the item table to get the confidence of each column field?

@Divya_Salve

Yes you can get he confidence of individual fields

Cheers

1 Like

That will not work for the table fields right? it will only give the regular fields confidence which is given the framework template

@Divya_Salve

Do you mean you want the table data that you extract …the confidence of that?

There you can get the confidence of whole table but not for individual columns I beleive

Cheers

I guess we can … as the confidence scores are present in the extracted results

@Divya_Salve

Okay so for individual columns if they are there…

Then once you have the table…

Dt.AsEnumerable.Where(function(x) x(“Fieldname”).ToString.Equals(“fieldnametosearch”)).Select(function(x) x(“Confidence”).ToString) can be used to get th ecofidence of each field

Here confidence…and fieldname are to be matched with the exact column names

Cheers

can I get the syntax to get the table along with confidence?
And if there are multiple line items we need to loop through each line item & get confidence

@Divya_Salve

Above syntax is to get the confidence of any field in table that you want …

Dt is the datatable in which thw table is stored

Where condition is applied on the column containing the fieldname…and checking if the value is equal to one of the fields fieldnametosearch(you can provide each fieldname and you would get confidence of eqch field…select is to select ghe confidence column value for the filtered field nametosearch

Hope this helps

Cheers

no the syntax to get the DT from extraction results is what I’m asking

@Divya_Salve

Try like this

ExtractedDataSet.Tables.Item(“Simple Fields”).Rows(0).Item(“requiredconfidence”).ToString

Here first we are getting the required table from the tables and then form the table we are extracting the required results

Idea again remains same you only need to change the columns

Cheers

THose are not simple fields right? I dont think this will work. Have you tried this & you got the confidence of each cell in table?

@Divya_Salve

I just gave made up column names and keys…you can check the exact keys and columns from
The locals panel

Cheers

Cheers