What is the recommended split of documents for training and evaluation, considering a total of 15 documents per vendor?
A. 12 documents for training the model, and 3 for evaluating the model.
B. 8 documents for training the model, and 7 for evaluating the model.
C. 7 documents for training the model, and 8 for evaluating the model.
D. 10 documents for training the model, and 5 for evaluating the model.
I think the Option A is the right answer here.
This split ensures that the model has enough data to learn effectively while still having a sufficient number of documents to evaluate its performance accurately.
There can’t be a exact number but just percentage. It’s a best practice to use 80% for training and remaining 20% for evaluation.
Refer this for more details.
Document Understanding - Training High Performing Models
@ashokkarale thank you for confirming this to me
This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.