-
## How to reproduce the behaviour
I try to call `python -m spacy convert ./ -t json -c iob` with this labeled NER training corpus:
[https://github.com/EuropeanaNewspapers/ner-corpora/blob/master/…
-
Using `alto2hocr.xsl` on [this alto file](https://digital.slub-dresden.de/data/kitodo/GottDie_453779263/GottDie_453779263_ocr/00000011.xml) via `ocr-fileformat` results in probably invalid hOCR since …
wrznr updated
5 years ago
-
@gwnet
CentOS7.6 kernel crash (kernel version 3.10.0-957.el7.x86_64)
use vdbench test performance for long time . kernel crash
dump dmsg
`
[ 0.000000] microcode: microcode updated ea…
-
I ran the below command and receive an error as below.
This happens in both python3.6.8 and python3.7.2
```
kraken -i bank.png bank.json segment --remove_hlines --no-script-detect --scale 2…
-
ocrd-tesserocr-segment-line does not give results for any of the files I tested. For example:
~~~sh
cd `mktemp -d`
wget https://ocr-d-repo.scc.kit.edu/api/v1/dataresources/8d8aa287-94ca-48e3-84a8…
-
The ocr library will often confuse 5 for S and Q for 0.
Create and use a custom .traineddata file specific to the level code font.
-
When running kraken segment I get some high quality bounding boxes. I am using a separate ocr to process these. Im trying to create an HOCR compatible page and the last piece of information is the pag…
-
I already asked [this question](http://stackoverflow.com/questions/40705203/tensorflow-label-image-recognize-so-slow?noredirect=1#comment68687966_40705203) on stackoverflow but It seems no one knows.
…
-
First of all, just to say hi and big thanks for your effort making this amazing software available for others to use.
I'm going to spend sometime converting this project to be more modern for large…
-
The experimental version exists before this commit 9753889ea266c3b8e412d77eb411617ec40d8393. It uses Tesseract (using pyocr). [ocropy](https://github.com/tmbdev/ocropy) looked promising the last time …