From the course: Azure AI Engineer Associate (AI-102) Cert Prep: Implement Knowledge Mining Solutions and Document Intelligence Solutions

Unlock the full course today

Join today to access over 24,900 courses taught by industry experts.

Implement a custom Document Intelligence model

Implement a custom Document Intelligence model

From the course: Azure AI Engineer Associate (AI-102) Cert Prep: Implement Knowledge Mining Solutions and Document Intelligence Solutions

Start my 1-month free trial Buy for my team

Implement a custom Document Intelligence model

“

- [Instructor] Let's have a look at the data we will be using to implement a custom Document Intelligence model. We have Globe Bank International, which has a form of investment product registration where people can detail their investment interest. It has personal details, selection marks, and check boxes that we would like to be extracted. We have stored this data in Azure blob storage. We will be implementing this custom model through the Document Intelligence Studio for ease of visualization and understanding. This can also be done programmatically if you so wish. Let's go through some of the model input requirements. We did cover this in the past section, but just as a reminder, custom models, the supported file types are PDF or images in JPEG, PNG, BMP, TIFF, or HEIF form. The minimum height of the text to be extracted is 12 pixels and the maximum number of pages for custom template models is 500, and for neural models is 50,000. For custom extraction model training, the total…

Contents