PLEASE DO NOT SHOUT - Sending messages in Large fontsize, RED color etc is not appreciated.
You have used a 0-zero instead of a CAPITAL O in your network spec, it should be O1c105 On Wednesday, March 28, 2018 at 12:24:02 PM UTC+5:30, [email protected] wrote: > > > > *Invalid network spec:01c105]* > *Missing ] at end of [Series]!* > *Failed to create network from spec: [1,0,0,1 Ct5,5,16 Mp3,3 Lfys64 Lfx128 > Lrx128 Lfx256 01c105]* > 2018년 3월 28일 수요일 오후 3시 53분 17초 UTC+9, [email protected] 님의 말: >> >> I type the command line in my computer ubuntu 16.04.03 LTS >> >> sudo lstmtraining --debug_interval -1 --traineddata >> /usr/share/tesseract-ocr/4.00/tessdata/kor.traineddata --net_spec* >> '[1,0,0,1 Ct5,5,16 Mp3,3 Lfys64 Lfx128 Lrx128 Lfx256 01c105]'* >> --train_listfile >> /usr/share/tesseract-ocr/4.00/tessdata/tesseract/training/trained_plus_chars_kor/kor.training_files.txt >> >> --eval_listfile >> /usr/share/tesseract-ocr/4.00/tessdata/tesseract/training/eval_plus_chars_kor/kor.training_files.txt >> >> --max_iterations 5000 >> >> >> I have an error . >> >> >> like >> >> >> Invalid network spec:01c105] >> Missing ] at end of [Series]! >> Failed to create network from spec: [1,0,0,1 Ct5,5,16 Mp3,3 Lfys64 Lfx128 >> Lrx128 Lfx256 01c105] >> >> >> But, I saw the wiki page >> >> https://github.com/tesseract-ocr/tesseract/wiki/VGSLSpecs >> >> >> Full Example: A 1-D LSTM capable of high quality OCR >> >> [1,1,0,48 Lbx256 O1c105] >> >> As layer descriptions: (Input layer is at the bottom, output at the top.) >> >> O1c105: Output layer produces 1-d (sequence) output, trained with CTC, >> outputting 105 classes. >> Lbx256: Bi-directional LSTM in x with 256 outputs >> 1,1,0,48: Input is a batch of 1 image of height 48 pixels in greyscale, >> treated >> as a 1-dimensional sequence of vertical pixel strips. >> []: The network is always expressed as a series of layers. >> >> This network works well for OCR, as long as the input image is carefully >> normalized in the vertical direction, with the baseline and meanline in >> constant places. >> >> <https://github.com/tesseract-ocr/tesseract/wiki/VGSLSpecs#full-example-a-multi-layer-lstm-capable-of-high-quality-ocr>Full >> >> Example: A multi-layer LSTM capable of high quality OCR >> >> *[1,0,0,1 Ct5,5,16 Mp3,3 Lfys64 Lfx128 Lrx128 Lfx256 O1c105]* >> >> As layer descriptions: (Input layer is at the bottom, output at the top.) >> >> O1c105: Output layer produces 1-d (sequence) output, trained with CTC, >> outputting 105 classes. >> Lfx256: Forward-only LSTM in x with 256 outputs >> Lrx128: Reverse-only LSTM in x with 128 outputs >> Lfx128: Forward-only LSTM in x with 128 outputs >> Lfys64: Dimension-summarizing LSTM, summarizing the y-dimension with 64 >> outputs >> >> >> Mp3,3: 3 x 3 Maxpool >> Ct5,5,16: 5 x 5 Convolution with 16 outputs and tanh non-linearity >> 1,0,0,1: Input is a batch of 1 image of variable size in greyscale*[]: The >> network is always expressed as a series of layers.* >> >> >> >> >> *I have no idea .. why I type [ ] these charcter put in there . Take place >> an error * >> >> >> *Could you help me .?? * >> >> -- 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 post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/e1b97153-13b9-40d6-b583-417a13ace47e%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
