See https://tesseract-ocr.github.io/tessdoc/tess4/TrainingTesseract-4.00.html#lstmtraining-command-line
Epoch has been recently added to the tesstrain makefile and converts to number of iterations based on amount of training data. On Wed, Apr 14, 2021, 01:36 GCP COGNEXT <gcp.cogn...@gmail.com> wrote: > Hi All, > > I would like to train (eng) tesseract model on 50k image files of a single > number and 50k corresponding .txt files (ground truth) for 1000 iterations. > > So, what does this mean *At iteration 300/400/1000*? I have the following > doubts. > > 1. > > What is iteration? Is it similar to epoch here? > 2. > > What does it mean here if I say 1000 iterations? Does it mean that all > my 50k files have been seen/trained by models 1000 times? Or does it mean > that only 1000 files from my 50k files have been seen by my model? > 3. > > Also please tell me what does it mean by this *At iteration > 300/400/1000*? What is 300, 400 & 1000 here? > > -- > 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-ocr+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/tesseract-ocr/e49626e0-0189-4832-89f6-8609d7b15389n%40googlegroups.com > <https://groups.google.com/d/msgid/tesseract-ocr/e49626e0-0189-4832-89f6-8609d7b15389n%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- 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-ocr+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/CAG2NduW7GcTC_rNVZAPpfoQgeCLR0SjeuQPAas%2Bi70FuOrHBEA%40mail.gmail.com.
