Opensoundscape

**Full Changelog**: https://github.com/kitzeslab/opensoundscape/compare/v0.7.1...v0.8.0

This release updates dependencies, fixes various bugs, updates documentation, and adds a few small features.

Notable feature additions
- Parsing metadata from AudioMoth firmware through 1.8.1 and from SongMeter Micros is now supported. The Audio.from_file method will automatically parse metadata and store it in the object's `.metadata` dictionary unless you specify `metadata=False`
- Audio.save() now saves metadata, so that you can retain or update .wav file metadata when using OpenSoundscape
- ResNet networks with pre-trained weights and a number of input channels other than 3 now average conv1 weights over the input channels

Dependencies
Python 3.9 no longer causes installation issues. This release updates OpenSoundscape's package dependencies in a way that allows Python 3.9 environments. Users with Apple Silicon chips to install OpenSoundscape hassle-free in a Python 3.9 environment (but may experience issues with Python 3.7 and 3.8).

If you encounter problems with installation, regardless of your platform, please open an issue.

What's Changed: a list of merged Pull Requests
* add scaling option to the Spectrogram class and MelSpectrogram class by zoharrpg in https://github.com/kitzeslab/opensoundscape/pull/505
* Annotation docs hotfix by rhine3 in https://github.com/kitzeslab/opensoundscape/pull/517
* Minor updates to annotation tutorial by rhine3 in https://github.com/kitzeslab/opensoundscape/pull/516
* Fixes Poetry Installer issue during CI by louisfh in https://github.com/kitzeslab/opensoundscape/pull/535
* change filepath of poetry by louisfh in https://github.com/kitzeslab/opensoundscape/pull/536
* Removed redundant empty load method from base model modules by jatinkhilnani in https://github.com/kitzeslab/opensoundscape/pull/537
* Update black version by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/539
* Issue 472 spectrogram methods by louisfh in https://github.com/kitzeslab/opensoundscape/pull/534
* Resolved - Resample with replacement 479 by jatinkhilnani in https://github.com/kitzeslab/opensoundscape/pull/497
* resolve 520 skip loading metadata by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/541
* resolve 513 Spectrogram loses properties on out-of-place method by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/542
* Fix loading of ResampleLoss CNN models by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/544
* Resolve Issue 507 metrics by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/540
* update pre-commit black version to 22.8.0 by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/551
* Issue 531 final clip by jatinkhilnani in https://github.com/kitzeslab/opensoundscape/pull/550
* Issue 496 tests for cnn arch by louisfh in https://github.com/kitzeslab/opensoundscape/pull/545
* different input channels should retain pre-trained weights by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/546
* Issue 523 AudioMoth 1.5-1.8.1 MetaData by syunkova in https://github.com/kitzeslab/opensoundscape/pull/553
* Sammlapp/issue391 by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/554
* v0.7.1 by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/557

New Contributors
* zoharrpg made their first contribution in https://github.com/kitzeslab/opensoundscape/pull/505
* syunkova made their first contribution in https://github.com/kitzeslab/opensoundscape/pull/553

**Full Changelog**: https://github.com/kitzeslab/opensoundscape/compare/v0.7.0...v0.7.1

New Contributors
* jatinkhilnani made their first contribution in https://github.com/kitzeslab/opensoundscape/pull/475

**Full Changelog**: https://github.com/kitzeslab/opensoundscape/compare/v0.6.2...v0.7.0

This release provides minor updates and bug fixes in the OpenSoundscape code base and documentation but does not make breaking changes to the API.

Merged Pull Requests:
* fix issue 438 cuda by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/454
* Issue 447 pandas dependency by louisfh in https://github.com/kitzeslab/opensoundscape/pull/461
* Resolves Issue 457 parse metadata by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/464
* modify make_clip_df to handle a single filepath string nicely by louisfh in https://github.com/kitzeslab/opensoundscape/pull/463
* Docs 0.6.2 issues 420 433 439 by louisfh in https://github.com/kitzeslab/opensoundscape/pull/465
* Docs for 0.6.2: issues 443 445 462 by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/466
* Issue 453 lstsq by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/470
* v0.6.2 by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/471


**Full Changelog**: https://github.com/kitzeslab/opensoundscape/compare/v0.6.1...v0.6.2

This release adds new functionality to OpenSoundscape and resolves a few issues.

Highlights
* Saving with model.save() and loading with `cnn.load_model()` or `cnn.load_outdated_model()`: Models are now saved and loaded as the entire object, so that the loaded model object has all attributes of the object when it was saved. Models saved with previous versions of opensoundscape must be loaded with `opensoundscape.torch.models.cnn.load_outdated_model` - see the tutorial "Predicting with pretrained CNNs" for details
* Directly load segments of audio files: using the new arguments `offset` and `duration` of Audio.from_file() you can quickly load an arbitrary section of a .WAV file without loading the whole file. (It works for mp3s too but is not as fast in that case.)
* Directly predict on "splits" of files without actually splitting them: New functionality including the ClipLoadingSpectrogramPreprocessor class allows you to quickly and easily generate CNN predictions on short (eg 5 second) segments of long audio files without needing to split the files into clips first. See "Predicting with pretrained CNNs" tutorial for an example
* Github homepage gives quick-start examples for tldr seekers
* model.device now allows you to move to a specific cuda device other than 'cuda:0'
* PytorchModel init allows user to specify name of architecture instead of architecture object, eg `PytorchModel('resnet50',classes=[0,1])`
* All models and architectures can be loaded without downloading pretrained weights. Resnet18Multiclass and Resnet18Binary have been refactored as subclasses of PytorchModel.

What's Changed
* add from_checkpoint() method to Resnet18Multiclass by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/413
* Add more informative Raven annotation file errors by rhine3 in https://github.com/kitzeslab/opensoundscape/pull/419
* Issue 410 cwtpeaks by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/416
* Handle user-specified devices like 'cuda:1' by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/422
* Issue 411 randominit by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/424
* Feat fast load by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/423
* Make split_and_save a method of Audio object by louisfh in https://github.com/kitzeslab/opensoundscape/pull/426
* Issue 427 load old models by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/428
* Allow models to be created using architecture name (string) by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/430
* update ML tutorials with new best practices and methods by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/431
* update init files to allow top-level import of opensoundscape by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/434
* add offset/duration example for Audio.from_file() by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/435
* Bug fixes, docs updates, and new prediction tools for 0.6.1 by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/432


**Full Changelog**: https://github.com/kitzeslab/opensoundscape/compare/v0.6.0...v0.6.1

This release represents a significant update to OpenSoundscape and includes changes to the API. It introduces new features and fixes various outstanding issues and bugs.

new feature highlights

- `annotations` module (replaces `raven` module) with new tools for manipulating audio annotations, such as those created in Raven software
- `signal` module for signal processing, including tool for Ruffed Grouse drumming detection
- directly load and predict on long audio files with CNNs without splitting first (splits audio into shorter clips in memory - this eliminates the need to split and save short clips for every long audio file before prediction) - see example in "Predicting with pre-trained CNNs" tutorial
- `Audio.from_file` now loads metadata from audio files into Audio object's `.metadata` dictionary. For files recorded by AudioMoths, this can include a recording timestamp and ambient temperature measurement, among other metadata
- `audiomoth` module assists in parsing AudioMoth recordings' metadata (e.g., ambient temperature and recording time) and file names (to datetime objects)

What's Changed: Pull Requests from Develop branch
* Bring develop "up to date" with master by rhine3 in https://github.com/kitzeslab/opensoundscape/pull/334
* Update contributor instructions by jkitzes in https://github.com/kitzeslab/opensoundscape/pull/356
* Issue 351: model training continuation by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/358
* Misc preprocessing updates by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/338
* Issue 364 dont resample by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/367
* Doc fixes by jkitzes in https://github.com/kitzeslab/opensoundscape/pull/374
* Feat reduced filesize by louisfh in https://github.com/kitzeslab/opensoundscape/pull/377
* Add ruffed grouse detection method by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/380
* Add functionality to split audio while predicting with CNN by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/381
* Add function for ruffed grouse detection by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/384
* Issue 287 moth tools by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/390
* Resolve a few preprocessing/training bugs by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/394
* annotation module by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/386
* Issue 378 ribbit bug by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/389
* Load metadata with Audio.from_file by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/398
* Remove species_table.py module by louisfh in https://github.com/kitzeslab/opensoundscape/pull/396
* Remove commands module by louisfh in https://github.com/kitzeslab/opensoundscape/pull/402
* Update docs modules and fix bugs by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/406
* Issue 294 per-class thresholds for binary predictions by louisfh in https://github.com/kitzeslab/opensoundscape/pull/399
* unpin dependencies and add pywavelets - all tests pass py=3.7 or 3.8 by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/408
* Version 0.6.0 by sammlapp in https://github.com/kitzeslab/opensoundscape/pull/409

New Contributors
* jkitzes made their first contribution in https://github.com/kitzeslab/opensoundscape/pull/356
* louisfh made their first contribution in https://github.com/kitzeslab/opensoundscape/pull/377

**Full Changelog**: https://github.com/kitzeslab/opensoundscape/compare/v0.5.0...v0.6.0