Crowsetta

Added
- Make `crowsetta.SimpleSeq.from_file` work with an "empty" csv file,
one that has no annotated segments
(e.g. because no audio was above threshold for segmenting)
[280](https://github.com/NickleDave/crowsetta/pull/280).
Fixes [264](https://github.com/NickleDave/crowsetta/issues/264).
- Add `default_label` argument to `crowsetta.SimpleSeq.from_file`
that will add labels to segments in a csv file if there are none
[280](https://github.com/NickleDave/crowsetta/pull/280).
Fixes [271](https://github.com/NickleDave/crowsetta/issues/271).
- Add example csv file from
[Jourjine et al. 2023 dataset](https://datadryad.org/stash/dataset/doi:10.5061/dryad.g79cnp5ts)
[280](https://github.com/NickleDave/crowsetta/pull/280).
Fixes [274](https://github.com/NickleDave/crowsetta/issues/274).
- Add how-to showing how to work with unannotated segmentation
in a csv file, using the csv from the Jourjine et al. 2023 dataset
[280](https://github.com/NickleDave/crowsetta/pull/280).
Fixes [275](https://github.com/NickleDave/crowsetta/issues/275).

Changed
- Rename `crowsetta.data` to `crowsetta.examples`,
simplify how `crowsetta.example` works
(to be more like `vocalpy.example`)
and give all the example annotation files
more specific names so that we can have
multiple examples per annotation format
[280](https://github.com/NickleDave/crowsetta/pull/280).
Fixes [278](https://github.com/NickleDave/crowsetta/issues/278).
- Import format classes at package level, so we can just type e.g.
`crowsetta.SimpleSeq` instead of `crowsetta.formats.seq.SimpleSeq`
("flat is better than nested")
[280](https://github.com/NickleDave/crowsetta/pull/280).
Fixes [273](https://github.com/NickleDave/crowsetta/issues/273).

Fixed
- Make `crowsetta.SimpleSeq.from_file` use `columns_map` arg
to rename only columns whose names are keys in the supplied dict,
and ignore other columns in the csv file
[280](https://github.com/NickleDave/crowsetta/pull/280).
Fixes [272](https://github.com/NickleDave/crowsetta/issues/272).

Fixed
- Replace deprecated `pandera.SchemaModel` with `DataFrameModel`,
fixes `AttributeError` on import after new install
[266](https://github.com/NickleDave/crowsetta/pull/266).
Fixes [265](https://github.com/NickleDave/crowsetta/issues/265).
- Change range of Python version supported from 3.9-3.11 to 3.10-3.12
[268](https://github.com/NickleDave/crowsetta/pull/268).
Fixes [267](https://github.com/NickleDave/crowsetta/issues/267).

Fixed
- Vendor code from evfuncs and birdsong-recognition-dataset packages,
to reduce the number of dependencies and make sure the code is maintained
[263](https://github.com/NickleDave/crowsetta/pull/263).
Fixes [262](https://github.com/NickleDave/crowsetta/issues/262).

Fixed
- Fix bug in "generic-seq" format; use validated dataframe
returned by pandera schema, so that "label" column is
coerced to strings
[258](https://github.com/NickleDave/crowsetta/pull/258).
Fixes [257](https://github.com/NickleDave/crowsetta/issues/257).

Added
- Add information on contributing and setting up a development environment
[212](https://github.com/NickleDave/crowsetta/pull/212).
Fixes [30](https://github.com/NickleDave/crowsetta/issues/30).
- Add method to convert generic sequence format to a pandas DataFrame
[216](https://github.com/NickleDave/crowsetta/pull/216).
- Add additional vignettes to docs:
on removing "silent" labels from TextGrid annotations,
on converting to the simple sequence and generic sequence formats
[216](https://github.com/NickleDave/crowsetta/pull/216).
Fixes [152](https://github.com/NickleDave/crowsetta/issues/152)
and [197](https://github.com/NickleDave/crowsetta/issues/197).
- Add format class for Audacity extended label track format
[226](https://github.com/NickleDave/crowsetta/pull/226).
Fixes [222](https://github.com/NickleDave/crowsetta/issues/222)
and [213](https://github.com/NickleDave/crowsetta/issues/213).
- Add the ability for a crowsetta.Annotation to have multiple sequences
[243](https://github.com/NickleDave/crowsetta/pull/243).
Fixes [42](https://github.com/NickleDave/crowsetta/issues/42).
- Rewrite TextGrid class to better handle file formats:
parse both "short" and default format in either UTF-8 or UTF-16
encoding; remove empty intervals from interval tiers by default;
can convert multiple interval tiers to a single crowsetta.Annotation
with multiple crowsetta.Sequences
[243](https://github.com/NickleDave/crowsetta/pull/243).
Fixes [241](https://github.com/NickleDave/crowsetta/issues/241)

Removed
- Remove `Segment.from_row` method, no longer used
[232](https://github.com/NickleDave/crowsetta/pull/232).
Fixes [231](https://github.com/NickleDave/crowsetta/issues/231)

Fixed
- Revise landing page of docs, and some vignettes.
Make other changes to clean up the docs build process
[216](https://github.com/NickleDave/crowsetta/pull/216).
- Coerce path-like attributes of `GenericSeq` dataframe schema to be strings.
This helps ensure these columns are always native Pandas types
[237](https://github.com/NickleDave/crowsetta/pull/237).
- Fix how the `crowsetta.Segment` class converts
onset sample and offset sample to int; correctly handle
multiple numpy integer subtypes
[238](https://github.com/NickleDave/crowsetta/pull/238).

Changed
- [c6ba100](https://github.com/vocalpy/crowsetta/commit/c6ba100d7335a880f2e1dbf66f5673ef562f3cc5)
Fix description and uri in pyproject.toml and crowsetta/__about__.py
- [f70828f](https://github.com/vocalpy/crowsetta/commit/f70828fc09102215ae9d8076f66a79d515db3388)
Make README images link to raw GitHub files so they render on PyPI