core
core copied to clipboard
Collection of OCR-related python tools and wrappers from @OCR-D
…to ensure that the given polygon is fully contained within that parent's polygon. Fixes #489. But please don't merge yet, I have another, better idea
We should track the runtime memory requirements besides runtime duration itself. And, if at all possible, the profile logger could also be fed with the image sizes (pixel resolution) of...
Add documentation for installation on ARM platforms (missing opencv-python, opencv-python-headless)
Installation on ARM is currently not possible because the required `opencv-python` is missing. Test result on Ubuntu 18.04.3 LTS (GNU/Linux 4.9.140-tegra aarch64): $ pip install ocrd Collecting ocrd Using cached...
The schema for the resource list defines a field `version_range` that should have the usual python syntax and should be used when downloading, esp. with wildcards.
In #278 we already had the case were the METS got broken due to ctrl+C in a single process. We discussed about parallelization but agreed to defer that topic. Now...
Here's the initial work on extending generateDS' GdsCollector to offer more helpful help messages. This replaces the buildin `parse/parseString` methods with variants that know about the filename (and passes it...
@mikegerber https://gitter.im/OCR-D/Lobby?at=5f92cffd6c8d484be2a2448e > > Could `ocrd-dummy` be aliased to `ocrd-copy`? I have had the need to just copy a file group from time to time, for example to copy a...
Despite the changes in the spec long ago, not requiring a processor to have an `output_file_grp`, core still enforces its defaults, both in the decorators… https://github.com/OCR-D/core/blob/6ca42f1614fbcaa07801d28d890d104223eb5aa3/ocrd/ocrd/decorators/ocrd_cli_options.py#L25 …and in the Processor...
Using any of core's API (even just `ocrd_utils`) now requires calling `initLogging` at some point prior to usage, or the respective subpackage's `getLogger` will complain with `CRITICAL`. So not only...
Currently with workspaces we can either keep images on the remote side by using http URLs in `mets:file/mets:FLocat/@xlink:href` (which means they have to be downloaded again and again during processing),...