Lstm training parameters

[Adriana Camilleri]

By any chance, is there any information out there about the training parameters that were used during the training of the models which are provided in the langdata_lstm directory, please?

Say, for the English model (eng.traineddata), what were the number of training iterations and what was the network topology specified as?

Thanks,

Adriana

[Shree Devi Kumar]

https://github.com/tesseract-ocr/tessdoc/blob/master/Data-Files-in-tessdata_fast.md

https://github.com/tesseract-ocr/tessdoc/blob/master/Data-Files-in-tessdata_best.md

--
You received this message because you are subscribed to the Google Groups "tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-oc...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/727a5dc3-17cb-4e9d-bc83-1b48b5b93f40n%40googlegroups.com.