diff options
| author | Nick White <git@njw.name> | 2021-06-22 15:27:51 +0100 |
|---|---|---|
| committer | Nick White <git@njw.name> | 2021-06-22 15:27:51 +0100 |
| commit | 99db57df8871bbf37c473d8eeddd0a1cf3de56ad (patch) | |
| tree | 69ee3bb70181189a1dc43d9dade6bcf7aaa146a3 /cmd | |
| parent | 714c55f077aceb982130642fc957db71f17007f0 (diff) | |
rescribe: Make it clearer that embedded training files are available to use
Diffstat (limited to 'cmd')
| -rw-r--r-- | cmd/rescribe/main.go | 13 |
1 files changed, 7 insertions, 6 deletions
diff --git a/cmd/rescribe/main.go b/cmd/rescribe/main.go index ea2590f..889a3d2 100644 --- a/cmd/rescribe/main.go +++ b/cmd/rescribe/main.go @@ -38,11 +38,6 @@ Process and OCR a book using the Rescribe pipeline on a local machine. OCR results are saved into the bookdir directory unless savedir is specified. - -Note that embedded Tesseract includes these training files: -- carolinemsv1_fast.traineddata (Caroline Miniscule) -- rescribefrav2_fast.traineddata (French historic printing) -- rescribev8_fast.traineddata (Latin historic printing) ` //go:embed tessdata.20210622.zip @@ -146,7 +141,13 @@ func main() { verbose := flag.Bool("v", false, "verbose") systess := flag.Bool("systess", false, "Use the system installed Tesseract, rather than the copy embedded in rescribe.") - training := flag.String("t", "trainings/rescribev8_fast.traineddata", "path to the tesseract training file to use") + training := flag.String("t", "rescribev8_fast.traineddata", `Path to the tesseract training file to use. +These training files are included in rescribe, and are always available: +- carolinemsv1_fast.traineddata (Caroline Miniscule) +- eng.traineddata (Modern English) +- rescribefrav2_fast.traineddata (French historic printing) +- rescribev8_fast.traineddata (Latin historic printing) + `) tesscmd := flag.String("tesscmd", deftesscmd, "The Tesseract executable to run. You may need to set this to the full path of Tesseract.exe if you're on Windows.") flag.Usage = func() { |