Issues
- 0
ketos segtrain not picking the best model?
#581 opened by johnlockejrr - 4
[feature request] Reproducible trainings
#580 opened by stweil - 6
Error on segmenting image (shapely-1.8.5)
#575 opened by johnlockejrr - 1
- 3
KeyError when calling get on any model in list
#576 opened by mbworth - 3
Cut a new release due to zenodo changes
#573 opened by IohannesArnold - 1
serialization / templates: prefer re-use of segment identifiers, if existing (from XML parser)
#568 opened by bertsky - 2
- 12
Kraken 4 does not ocr htr
#571 opened by Hoekman57 - 4
How to create training dataset programmatically?
#562 opened by hhuseyinpay - 0
- 4
recognizer: Shapely problem, unspecific error message
#565 opened by bertsky - 3
regression: no model for tags type=default
#564 opened by bertsky - 5
"kraken get 10.5281/zenodo.2577813" and "kraken list" commands failing after successful kraken==4.3.13 install in WSL2
#561 opened by Blu5Morpheus - 2
Is it necessary for `kraken.binarization.nlbin` to raise an error on blank images ?
#554 opened by sven-nm - 2
Word error rate added in your release 4.3.10
#559 opened by soniasol - 8
`KeyError` with `kraken list/get/show`
#550 opened by csidirop - 2
suboptimal model saved
#539 opened by dstoekl - 7
- 2
- 11
- 0
Fix linting issue
#553 opened by slobodan-ilic - 9
Error initiating recognition training, kraken 4.13.3
#552 opened by hlapin - 10
- 2
UnpicklingError when running ketos rotrain
#548 opened by Svetlana-Yatsyk - 8
improve polygons of bottom lines
#544 opened by dstoekl - 2
tags required or not?
#512 opened by bertsky - 7
Training with bounding box data (png/gt.txt pair)
#516 opened by rohanchn - 3
Colon in target label for --merge-regions
#532 opened by rohanchn - 3
- 11
- 0
- 2
Suppress warnings from coremltools?
#536 opened by stweil - 3
Website Redirect
#533 opened by waynegraham - 0
ketos test gives different result than validation in training for the same model and same dataset
#540 opened by dstoekl - 0
- 3
Python 3.8 support
#537 opened by bertsky - 2
torch version compatibility
#515 opened by bertsky - 5
- 4
- 2
maximum recursion during a test
#530 opened by lauxley - 7
Help with VGSL syntax
#527 opened by lamaeldo - 4
- 13
Transcription model producing empty output
#523 opened by mabarber92 - 12
- 2
Current Forced Alignement does not work when the model is trained on normalized data and not the transcription
#521 opened by PonteIneptique - 2
Stop training early if accuracy is 100%
#519 opened by stweil - 2
crappy segmentation
#507 opened by particitae - 7
Training on large binary dataset
#509 opened by colibrisson - 3
Optimisation issue
#508 opened by colibrisson