thank you shree for the answers. It was actually a misunderstanding on my part. I thought there is no way it would reach 0 error when on the training log, it doesnt even reach 0. Where does this disparity come from?
On Thu, 16 Apr 2020 at 15:43, Shree Devi Kumar <[email protected]> wrote: > --verbosity Amount of diagnostic information to output (0-2). (type:int > default:1) > > Try with verbosity 2 - it should show even when the ground truth and OCR > text are same. > > I think your model is correctly recognizing all lines it was trained on > but not on the eval lines. > > It should also show you the accuracy percentage at end. > > I am assuming that the model is correctly recognizing the lines it was tr > > > > On Thu, Apr 16, 2020 at 7:42 AM Usamah Jundi <[email protected]> > wrote: > >> Hello shree, i did not specify any verbosity level. The same exact >> command with the eval list argument pointing to other lists works just fine >> >> On Wednesday, April 15, 2020 at 6:53:19 PM UTC+7, shree wrote: >>> >>> lstmeval has different verbosity levels. Which one did you use? >>> >>> On Wed, Apr 15, 2020 at 4:17 PM Usamah Jundi <[email protected]> wrote: >>> >>>> Hi, sorry for the brief title, let me explain my situation a bit. >>>> >>>> So what i've done is: >>>> >>>> 1. Use tesstrain.sh to generate the training files (the.lstmf, .txt and >>>> that one other format) >>>> 2. Fine-Tune the best model to said training files >>>> 3. Perform lstmeval using one of the resulting checkpoints on the same >>>> file i trained on >>>> >>>> Expected result : >>>> - the program loads every files listed on the .txt generated in (1), >>>> and performs validation, and in doing so prints out the Truth and OCR of >>>> every sample >>>> >>>> What i got: >>>> - the program prints the file names in the .txt generated in (1), but >>>> did not perform validation, no Truth and OCR result being written seems to >>>> indicate it fails to read the lines? >>>> >>>> Things to note: >>>> - other checkpoints trained on the same data does not work, too >>>> - it works when i changed data list (not the one i use for training) or >>>> use the .lstm file you extract from off-the-shelf .traineddata files >>>> >>>> Is this intended behaviour? or am i doing something wrong? >>>> >>>> -- >>>> 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 [email protected]. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/tesseract-ocr/4b75e0be-1bdc-4c1c-a108-8055aefc737b%40googlegroups.com >>>> <https://groups.google.com/d/msgid/tesseract-ocr/4b75e0be-1bdc-4c1c-a108-8055aefc737b%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>> >>> >>> -- >>> >>> ____________________________________________________________ >>> भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com >>> >> -- >> 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 [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/tesseract-ocr/3975d6ad-4782-4f0b-aee2-e4aff8ed7749%40googlegroups.com >> <https://groups.google.com/d/msgid/tesseract-ocr/3975d6ad-4782-4f0b-aee2-e4aff8ed7749%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> > > > -- > > ____________________________________________________________ > भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com > > -- > 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 [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/tesseract-ocr/CAG2NduXe9CxbL16vatkNByb__me%2BE%2BJz4q8ddgUxzY2bp0-bhA%40mail.gmail.com > <https://groups.google.com/d/msgid/tesseract-ocr/CAG2NduXe9CxbL16vatkNByb__me%2BE%2BJz4q8ddgUxzY2bp0-bhA%40mail.gmail.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 [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/CAKH1bLN1nzRnxecdZ9F6mfpmmusEVSDEFJk3CjDEHdp7dJOyKg%40mail.gmail.com.
