Document Understanding document not detetcted

Help Activities
, , ,
1

I have Data Labelled for a particular project with 35 documents.
All the documents fields were identified correctly while Labelling.
But for few sismilar documents , no data is getting identified in Data Extraction part.
Need some assistance.

2

Can you check the both the documents having any difference .

Note : - if both are different then you need to train the those documents also

@Ritaman_Baral

3

exact document with whoich I did Data labelling

4

if possible can you share the both the sample documents .

@Ritaman_Baral

5

not possible…bot the documents are same…I am working in client machine

6

Hi @Ritaman_Baral ,

Could you let us know some more details of the Data samples :

  1. Are all the Document samples Digital or mixture of Both Digital and Scanned.
  2. How many different templates are present for the Document type used assuming you have only one type of document to be processed.
  3. What is the OCR Engine used ?
  4. How many fields are to be extracted from the Document ?
7
  1. Mixed
  2. 2-3
  3. OmniPage
  4. 8
8

@Ritaman_Baral ,

From the Provided data above, it is recommended to have at least 40 - 80 Pages of data for your dataset samples.

Where the dataset is evenly distributed (balanced) for different templates.

image
image606×568 24.9 KB

9

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