Data Extraction (Data extraction scope --> Machine learning extorter --> configure Extractors)

I try to Extract Details From the PDF I use the Document understanding for the extraction in the taxonomy (Data extraction scope → Machine learning extorter → configure Extractors)

in the Configure Extractor there are some fields only there
I want to extract the(Supplier Code, Item code and purchase ID) But those fields are Not available in the Data extraction scope (Configure Extraction )

how to extract the details of Supplier Code / Item Code / Purchase ID

From Extractor not applicable for my process.


Try using po number or item po number…they might be linked to the required ids


Hi @krishna_priya

You might have to use data labeling session and train a custom ML model in order to extract the specified custom fields such as Supplier Code from your invoices, as the Out-of-the-Box Invoice model may not contain the fields what you want to extract.

Best Regards.

yeah I tried bro
But Data can’t extracted

this is any is there for extract the details of (Supplier Code / Item Code / Purchase ID)

want to extract the details of the Supplier Code / Item Code / Purchase ID in the PDF


Then you need to retrain the model with the columns you need deploy a new skill



Yes. Already available Out-of-the-Box model may not contain such fields in their schema already. You need to use a feature called ‘Transfer Learning’ in order to train your model to extract such custom fields.

In this you will take an existing model & add the fields that you want to extract. Then train the model with sufficient data, train & transfer the model, which can be used to extract the fields.

In order to do this, you need to create a dataset in the action center, label the data field that you want to extract, export the labelled data and train it in the pipeline. Deploy the ML skill on the trained pipenline package and use the same in the project.

Data labeling would look something like this:

Hope this helps,
Best Regards.


I’m using the Endpoint and Document Understating ApiKey in the machine learning extraction

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.