tesstrain

Training workflow for Tesseract 4 as a Makefile for dependency tracking and building the required software from source.

Install

leptonica, tesseract

You will need a recent version (>= 4.0.0beta1) of tesseract built with the training tools and matching leptonica bindings. Build instructions and more can be found in the Tesseract project wiki.

Alternatively, you can build leptonica and tesseract within this project and install it to a subdirectory ./usr in the repo:

  make leptonica tesseract

Tesseract will be built from the git repository, which requires CMake, autotools (including autotools-archive) and some additional libraries for the training tools. See the installation notes in the tesseract repository.

Python

You need a recent version of Python 3.x. For image processing the Python library Pillow is used. If you don't have a global installation, please use the provided requirements file pip install -r requirements.txt.

Choose model name

Choose a name for your model. By convention, Tesseract stack models including language-specific resources use (lowercase) three-letter codes defined in ISO 639 with additional information separated by underscore. E.g., chi_tra_vert for traditional Chinese with vertical typesetting. Language-independent (i.e. script-specific) models use the capitalized name of the script type as identifier. E.g., Hangul_vert for Hangul script with vertical typesetting. In the following, the model name is referenced by MODEL_NAME.

Provide ground truth

Place ground truth consisting of line images and transcriptions in the folder data/MODEL_NAME-ground-truth. This list of files will be split into training and evaluation data, the ratio is defined by the RATIO_TRAIN variable.

Images must be TIFF and have the extension .tif or PNG and have the extension .png, .bin.png or .nrm.png.

Transcriptions must be single-line plain text and have the same name as the line image but with the image extension replaced by .gt.txt.

The repository contains a ZIP archive with sample ground truth, see ocrd-testset.zip. Extract it to ./data/foo-ground-truth and run make training.

NOTE: If you want to generate line images for transcription from a full page, see tips in issue 7 and in particular @Shreeshrii's shell script.

Train

 make training MODEL_NAME=name-of-the-resulting-model

which is basically a shortcut for

   make unicharset lists proto-model training

Run make help to see all the possible targets and variables:


  Targets

    unicharset       Create unicharset
    lists            Create lists of lstmf filenames for training and eval
    training         Start training
    traineddata      Create best and fast .traineddata files from each .checkpoint file
    proto-model      Build the proto model
    leptonica        Build leptonica
    tesseract        Build tesseract
    tesseract-langs  Download tesseract-langs
    clean            Clean all generated files

  Variables

    MODEL_NAME         Name of the model to be built. Default: foo
    START_MODEL        Name of the model to continue from. Default: ''
    PROTO_MODEL        Name of the proto model. Default: 'data/foo/foo.traineddata'
    CORES              No of cores to use for compiling leptonica/tesseract. Default: 4
    LEPTONICA_VERSION  Leptonica version. Default: 1.78.0
    TESSERACT_VERSION  Tesseract commit. Default: 4.1.0
    TESSDATA_REPO      Tesseract model repo to use. Default: _best
    TESSDATA           Path to the .traineddata directory to start finetuning from. Default: ./usr/share/tessdata
    GROUND_TRUTH_DIR   Ground truth directory. Default: data/MODEL_NAME-ground-truth
    OUTPUT_DIR         Output directory for generated files. Default: data/MODEL_NAME
    MAX_ITERATIONS     Max iterations. Default: 10000
    NET_SPEC           Network specification. Default: [1,36,0,1 Ct3,3,16 Mp3,3 Lfys48 Lfx96 Lrx96 Lfx256 O1c\#\#\#]
    FINETUNE_TYPE      Finetune Training Type - Impact, Plus, Layer or blank. Default: ''
    LANG_TYPE          Language Type - Indic, RTL or blank. Default: ''
    PSM                Page segmentation mode. Default: 6
    RANDOM_SEED        Random seed for shuffling of the training data. Default: 0
    RATIO_TRAIN        Ratio of train / eval training data. Default: 0.90

Make model files (traineddata)

When the training is finished, it will write a traineddata file which can be used for text recognition with Tesseract. Note that this file does not include a dictionary. The tesseract executable therefore prints an warning.

It is also possible to create additional traineddata files from intermediate training results (the so called checkpoints). This can even be done while the training is still running. Example:

# Add MODEL_NAME and OUTPUT_DIR like for the training.
make traineddata

This will create two directories tessdata_best and tessdata_fast in OUTPUT_DIR with a best (double based) and fast (int based) model for each checkpoint.

It is also possible to create models for selected checkpoints only. Examples:

# Make traineddata for the checkpoint files of the last three weeks.
make traineddata CHECKPOINT_FILES="$(find data/foo -name '*.checkpoint' -mtime -21)"

# Make traineddata for the last two checkpoint files.
make traineddata CHECKPOINT_FILES="$(ls -t data/foo/checkpoints/*.checkpoint | head -2)"

# Make traineddata for all checkpoint files with CER better than 1 %.
make traineddata CHECKPOINT_FILES="$(ls data/foo/checkpoints/*[^1-9]0.*.checkpoint)"

Add MODEL_NAME and OUTPUT_DIR and replace data/foo by the output directory if needed.

License

Software is provided under the terms of the Apache 2.0 license.

Sample training data provided by Deutsches Textarchiv is in the public domain.

xhuvom / tesstrain