Robert Sachunsky

OcrdExif / page_from_image: show/use orientation

1

I don't know how realistic this is for OCR-D data providers: What if the original images are not oriented upright / positive, and the imaging system already knows about this...

Now that we have support for pageId filtering and regexes in `find` / `remove` / `remove-group`, how about support for **partial cloning** with a filter (either comma-separated or regex) on...

workspace download: also traverse dependent file groups?

3

When I want to download a PAGE-XML from remote, it would be very helpful if core would also download all the files referenced in `/PcGts/Page/@imageFilename` and `*/AlternativeImage/@filename`. Is this feasible?

ocrd_utils.coordinates_for_segment: clip to parent?

5

We have the utility function `coordinates_for_segment` for conversion of polygons from some derived image and its coordinate transform to absolute coordinates of the original image. This is all the function...

I know this is already [on the agenda](https://github.com/OCR-D/core/blob/76ea92710a72db278bc38fd7b698ab89b77580d1/ocrd/ocrd/workspace_bagger.py#L66), but it should be more visible IMHO: `ocrd zip bag` could really use an option that allows filtering (negatively) or selecting (positively)...

support pre-training or fine-tuning schemes

18

Thanks for sharing your work, it's awesome! I am eager to train this on my own materials, but they are comparatively scarce and I don't have the computational capacities to...

fix grouping GT vs prediction by not assuming only pred have score

9

This removes the assumption that only the prediction dataset contains `score` values.

CI: make forks work and enable for PRs

1

@i008 Wow, I did not notice you already set up a CI! Alas, it still fails in my fork due to permissions: ``` Run docker/login-action@v1 Error: Username and password required...

enable caching

1

I wonder if browsing can be speeded up by enabling `@st.cache` for `inspector.visualize_image`. I have tried with various options (`allow_output_mutation=True`, `persist=True`, even various `hash_funcs=` trying to avoid hashing certain types)....

I am trying to make sense of the scores output provided here (which puzzles me over what I am used to from `pycocotools` itself, even more so after I have...