Thank you Once again for creating content onDU. I wish you could have explained, how you created regex for the Invoice 1 and Invoice 2 fields instead of showing regex values for receipt fields since it was easy. I tried using chat GPT. it didn't work for the date since one is a native pdf and the other scanned PDF. The format of strDocText is different for both. I had to use your regex expressions given in the description, not sure how you created them.
Thanks for your comment For regex based , we need to have a prior knowledge of regex and that’s a major drawback for using This . My idea here is to show you the challenge what happens when you use regex … and you are correct , it’s difficult… now coming to solution- the answer is to use the intelligent keyword classifier and it’s training , please find the videos in this same playlist
Hi Mukesh Can you please do the video on CMS1500 form using DU in market place already we have solution but how to use that template can you please explain briefly
This is the Out-of-the-box Pre-trained ML Packages in UiPath Document Understanding , You would need to use the AI center to deploy this packages and then use the same in your automation - Begin by creating a dataset that contains the documents you want to process. - You can use the Document Manager to import, store, and categorize these documents. - Navigate to ML Packages in your UiPath environment. - Choose the Out-of-the-box Packages option. - From the list of available pre-trained extractors, choose the one that aligns with your document type or use case. - Provide the necessary details - Depending on your workflow type, you can integrate the extractors in different ways: - Classic RPA Workflows (Studio Desktop): - Cross-Platform RPA Workflows (Studio Desktop or Studio Web):
Hi Mukesh, I have seen in the classifiers and extractors there are the following: machine learning classifier and machine learning extractor, have you tried those functionalities?
Once the document is classified, MyValidatedClassificationResult holds the classification results, which are then used to direct the workflow to the appropriate extraction process for that document type. For example, if a document is classified as an invoice, the workflow will proceed with the steps required to extract data specific to invoices (such as invoice numbers, dates, amounts, etc.).
@@MukeshKala In the "Data Extraction Scope," we use "ClassificationResult[0].DocumentTypeId," not "MyValidatedClassificationResult". What is the use of "MyValidatedClassificationResult" if the document is classifying with "ClassificationResult[0].DocumentTypeId"?
extraction completed
Awesome
Sir Can you please make series on Document Understanding module present in AI Centre
Yes once we complete the basic one , then we move to the ML
Thank you Once again for creating content onDU. I wish you could have explained, how you created regex for the Invoice 1 and Invoice 2 fields instead of showing regex values for receipt fields since it was easy. I tried using chat GPT. it didn't work for the date since one is a native pdf and the other scanned PDF. The format of strDocText is different for both. I had to use your regex expressions given in the description, not sure how you created them.
Thanks for your comment
For regex based , we need to have a prior knowledge of regex and that’s a major drawback for using This . My idea here is to show you the challenge what happens when you use regex … and you are correct , it’s difficult… now coming to solution- the answer is to use the intelligent keyword classifier and it’s training , please find the videos in this same playlist
@@MukeshKala Thanks Mukesh for the response. I’m still going through the playlist.
Hi Mukesh
Can you please do the video on CMS1500 form using DU in market place already we have solution but how to use that template can you please explain briefly
This is the Out-of-the-box Pre-trained ML Packages in UiPath Document Understanding , You would need to use the AI center to deploy this packages and then use the same in your automation
- Begin by creating a dataset that contains the documents you want to process.
- You can use the Document Manager to import, store, and categorize these documents.
- Navigate to ML Packages in your UiPath environment.
- Choose the Out-of-the-box Packages option.
- From the list of available pre-trained extractors, choose the one that aligns with your document type or use case.
- Provide the necessary details
- Depending on your workflow type, you can integrate the extractors in different ways:
- Classic RPA Workflows (Studio Desktop):
- Cross-Platform RPA Workflows (Studio Desktop or Studio Web):
Extraction completed,thank you.
Glad to hear that , Thanks
In version 2024.4 there is a new extraction mode "Generative Extrator"
Yes these are the new set of activities the DU is having
Extraction Success
That’s awesome 🙌
Extraction Successful
Glad to hear that , Thanks for posting
Sir, how did you create the regex. Can you tell us the tool you have used so we can make it ourselves?
It’s explained in video
16:57 : How does regex work
27:47 : How to get regex online
Yeah done...Great Mukesh. Thankyou :)
Glad to hear that you were able to follow along
😊@@MukeshKala
Hi Mukesh, I have seen in the classifiers and extractors there are the following: machine learning classifier and machine learning extractor, have you tried those functionalities?
Yes , Once we come to AI Center we will cover those
@@MukeshKala Genial!
Extraction Succesful
Wow , Thanks for sharing
Hi Mukesh
Why do we need the variable "MyValidatedClassificationResult", which we got from the "Train Classifiers Scope"?
Once the document is classified, MyValidatedClassificationResult holds the classification results, which are then used to direct the workflow to the appropriate extraction process for that document type.
For example, if a document is classified as an invoice, the workflow will proceed with the steps required to extract data specific to invoices (such as invoice numbers, dates, amounts, etc.).
@@MukeshKala In the "Data Extraction Scope," we use "ClassificationResult[0].DocumentTypeId," not "MyValidatedClassificationResult". What is the use of "MyValidatedClassificationResult" if the document is classifying with "ClassificationResult[0].DocumentTypeId"?
Great!
Hope you were able to follow along
@@MukeshKala Yes
Done with Extraction
Awesome 🙌
Extraction completed Sir 🫡
Great 😊
From where can I get the documents?
It’s all uploaded on my website … find the link in description, you can watch the initial videos to build the code along with