Repeating words distribution in train set #12071
Unanswered
pooja-kabra
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I am finetuning the text recognition network on product labels. Some text fields occur repeatedly on the labels. For example, PRODUCT_ID: 343765. All labels will have the text 'PRODUCT_ID'. How should the word distribution be in the training set for these texts? For example in a 5k dataset, how many samples should these repeating words take up?
Beta Was this translation helpful? Give feedback.
All reactions