…les.

…ed labels (model output) to more verbose format usable by `export_music.py`.

… to PageLayout.

…eXML or ALTO.

…ALTO.

… loading.

… + loading.

…to define settings and `page_parser.py` to create music exporter object of `music/export_music/ExportMusicPage`.

…g box around polygon.

…t.py`. Get names only from Yolo `result.names`.

…f lines.

…music.py` a stand-alone script.

… text Layout engine to work only with 'text' lines.

…h with its own setting and set of categories to work with.

# Conflicts:
#	pero_ocr/core/layout.py

…tructure.

…ltoxml. part 2.

- in layout_helpers.py add searching for a region by ID
-  in page_parser.py change log error (missing crop) back to raise exception
-  in page_parser.py change naming convention for config section names to {section_name}_\d+. All following examples are possible (with warnings) for OCR section: OCR, OCR_0, OCR_0_asdf...

…ategories instead of json string.

- `image_size` can be either int or (int, int) according to ultralytics docu: https://docs.ultralytics.com/modes/predict/#inference-arguments

…ort to page_xml region custom.

- `TransformerEngineLineOCR` now translates all outputs according to `music_dictionary` dictionary in ocr.json (or omr.json), if exists.
- Move `MusicTranslator` to a separate file for nicer imports everywhere.

…on `sort_regions_in_reading_order`

- print more informative warning message
- add `__init__.py` to enable python package functionality for future.

- delete `layout.py/PageLayout.get_regions_of_category` as its functionality was substituted by `layout_helpers.py/split_page_layout_by_categories`. For sorting regions use (Naive|Smart)RegionSorter during `parse_folder` execution resulting in regions being sorted in xml output.
- in `music_exporter.py` make translator optional argument.

- `LINE_CATEGORIES`: for detected region also create a TextLine for future transcription.
- `CATEGORIES`: categories to save in `LayoutExtractorYolo`, delete others. (if `None` or `[]`, save all)
- disable YOLO detection results printing
- filter categories for `PageDecoder`

…verything around it.

- substitution dictionary needs reversed format now. Translate key to values (to be more obvious). Example follows:
- SSemantic music, Model output (now keys of dictionary): >2 + kGM + B3z + C4z + |..."
- Semantic music (now values of dictionary): clef-G2 + keySignature-GM + note-B3_eighth + note-C4_eighth + barline...")

- enable by adding `--output-render-category` argument

…music.py`.

… TextLine if new confidence is higher.

- move `output_substitution` to PageOCR after confidence estimation.
- add `TextLine.get_labels` for easier confidence estimation.
- add `get_line_confidence_median` to save literally one line of code later...

… estimation.

Problem was that confidence was estimated using `confidence_estimation.py/get_line_confidence` without cutting `log_probs` according to `line.logit_coords`. New method solves it.

…t was moved to ocr/omr.json (OCR engine config jon).

…inal symbol if not found in dictionary.

…es to `PageOCR` config section.

- `SUBSTITUTE_OUTPUT` ('yes' or 'no'): enables (or disables) output substitution (enabled by default)
- `SUBSTITUTE_OUTPUT_ATOMIC` ('yes' or 'no'): if 'yes' and any symbol cannot be translated, return original line. if 'no', line may be partially translated.

…dual lines.

Parameter sets if PageOCR should update to new line:
 - every time (false)
 - only if better confidence (true)

Applies in case of rerunning OCR on previously transcribed line)

- Versions older than 4.2 defines baseline as a simple float. (that's where the original baseline comes from)
- version 4.2 and never defines baseline as a PointsType string with recommend format: "x1,y1 x2,y2 ..."

…port options.

- Versions older than 4.2 defines baseline as a simple float. (baseline is exported as mean of all Y baseline points)
- version 4.2 and never defines baseline as a PointsType string with recommend format: "x1,y1 x2,y2 ..."

…on3.9)

…n Python3.9)

Old XMLs on input don't have category => line.category = None, OCR (and others) have to be set to `[]` by default to process ALL PAGES.

1) Remove `ultralytics` and `music21` from dependencies for the whole projest. the user will have to install them when needed.
2) Import `ultralytics` only when needed, so it doesn't create import error for specific numpy versions.

Ultralytics has this dependency right now: "numpy>=1.23.5,<2.0.0". See current at [github.com/ultralytics/ultralytics/blob/main/pyproject.toml](https://github.com/ultralytics/ultralytics/blob/69cfc8aa228dbf1267975f82fcae9a24665f23b9/pyproject.toml#L67)

…t away.

In `smart_sorter.py`:
- if less then to engines filtered, return original page_layout and not only the split one.

In `music structures.py`:
- change type of `lengths` to numpy array, fix min_length to take from numbers and not names.
- ensure `encoded_group` is not None before appending it to the voice.

full comment: [pero-ocr/pull/56/#issuecomment-2245202776](#56)

…whole region to positive or negative (ignore categories of lines inside the region)

Export multirest as a simple default 'whole' rest.

…on category, None = 'text')

…fidence -- in case when there are no logits (i.e. logits.shape[0] == 0) the confidence cannot be calculated.