Seeking Guidance for Invoice Extraction Project Using ML.NET with Blazor

Dinushan 0 Reputation points
2024-03-22T09:39:56.1966667+00:00

Hello ML.NET community,

I am embarking on a project to develop an invoice extraction application using Blazor, and I'm planning to integrate ML.NET for accurate invoice data extraction. As a newcomer to both Blazor and ML.NET, I am seeking guidance and advice from experienced members of the community.

Specifically, I would appreciate assistance with the following areas:

  1. Integrating ML.NET into a Blazor application for invoice extraction tasks.
  2. Best practices for handling PDF files within a Blazor application and preprocessing invoice images.
  3. Strategies for training an ML.NET model within the Blazor framework and ensuring accuracy.
  4. Recommendations for tools, libraries, or approaches to efficiently handle different invoice formats and variations.

If anyone has experience or insights related to using ML.NET with Blazor or similar technologies, I would be grateful for any tips, code snippets, tutorials, or recommended resources.

Thank you for your time and support. I look forward to learning from the community and making progress on this exciting project.

Best regards, [Your Name]

Developer technologies | .NET | Blazor
Developer technologies | .NET | .NET Machine Learning
{count} votes

2 answers

Sort by: Most helpful
  1. Olaf Helper 47,516 Reputation points
    2024-03-22T11:28:53.0733333+00:00

    Have look at Azure AI Document Intelligence, may a better option.

    1 person found this answer helpful.

  2. Bruce (SqlWork.com) 78,316 Reputation points Volunteer Moderator
    2024-03-30T22:55:39.2533333+00:00

    Unless you are familiar which ocr and ai models, you will probable want a 3rd party solution. Due to model size and model proprietary, most will be online solutions similar to the suggested azure solution.

    if you want open source solution, most will be python based. You could use these to create a webapi called by your blazor app.

    0 comments No comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.