This set of traineddata files has support for the legacy recognizer with –oem 0 and for LSTM models with –oem 1. Tessdata tagged 4.0.0 has the models from Sept 2017 that have been updated with Integer versions of tessdata_best LSTM models. Data Files for Version 4.00 (November 29, 2016) The legacy engine is not supported with these files, so Tesseract’s oem modes ‘0’ and ‘2’ won’t work with them. Note: When using the new models in the tessdata_best and tessdata_fast repositories, only the new LSTM-based OCR engine is supported. The current set of files in tessdata have the legacy models and newer LSTM models (integer versions of 4.00.00 alpha models in tessdata_best). The 4.00 files from November 2016 have both legacy and older LSTM models. The third set in tessdata is the only one that supports the legacy recognizer. The only set of files which can be used for certain retraining scenarios for advanced users. Tessdata_best is for people willing to trade a lot of speed for slightly better accuracy. Most users will want tessdata_fast and that is what will be shipped as part of Linux distributions. Integerized LSTM of a smaller network than tessdata-best Slightly less accurate than tessdata-best Legacy + LSTM (integerized tessdata-best) These are compatible with Tesseract 4.0x + and 5.0.0.Alpha. traineddata files on GitHub in three separate repositories. osd is compatible with version 3.01 and up, and equ is compatible with version 3.02 and up. Note: These two data files are compatible with older versions of Tesseract. The legacy tesseract engine (–oem 0) is NOT supported with these files, so Tesseract’s oem modes ‘0’ and ‘2’ won’t work with them. When using the traineddata files from the tessdata_best and tessdata_fast repositories, only the new LSTM-based OCR engine (–oem 1) is supported.
(Cube based legacy tesseract models for Hindi, Arabic etc. The LSTM models have been updated with Integer version of tessdata_best LSTM models. tessdata (Nov 2016 and Sep 2017) These have legacy tesseract models from 2016.
These are the only models that can be used as base for finetune training. tessdata_best (Sep 2017) best results on Google’s eval data, slower, Float models.tessdata_fast (Sep 2017) best “value for money” in speed vs accuracy, Integer models.These are made available in three separate repositories. traineddata files trained at Google, for tesseract versions 4.00 and above. Traineddata Files for Version 4.00 + Tesseract documentation View on GitHub Traineddata Files for Version 4.00 + Traineddata Files for Version 4.00 + | tessdoc Skip to the content.
0 kommentar(er)
