1.1.1

installation

• Set minimum python version to 3.10, include py312 in test suite, replace deprecated pkg_resources with importlib.metadata [#73]
• Set minimum tensorflow version to 2.16.1 [#73]
• Pre-trained neural networks updated for compatibility with Keras 3 [#73]
• Dockerfile images now use debian bookworm [#73]

1.1.1rc1

installation

• Set minimum python version to 3.10, include py312 in test suite, replace deprecated pkg_resources with importlib.metadata [#73]
• Set minimum tensorflow version to 2.16.1 [#73]
• Pre-trained neural networks updated for compatibility with Keras 3 [#73]
• Dockerfile images now use debian bookworm [#73]

1.1.0

new features

• builder.trained_networks jwst_cal.zip includes updated (v2) img3_reg and new spec3_reg predictive models for image and spectroscopic data [#58]
• preprocessor.ingest.JwstCalIngest class and cmdline script for automated training data ingest [#57]
• extractor.radio.JwstCalRadio subclass for scraping datasets from MAST using ASN metadata [#51]
• extractor.scrape.FitsScraper.scrape_dataframe method added for scraping Fits data from dataframe [#52]

enhancements

• skopes.jwst.cal.predict generates predictions for spectrosopic datasets in addition to image data. This update also allows further customization of user arguments: [#58]
  • obs to specify selection of a program ID + observation number
  • input_path accepts either a directory (default) or a filename. If filename, the script will try to find any input exposures that belong to the same program and observation number as that file.
  • sfx attribute is now customizable on instantiation of the class object (default is '_uncal.fits')
• architect.builder.Builder.save_model uses preferred keras archive format by default [#50]
• preprocessor.transform.SkyTransformer set offsets to 0 for gs/targ fiducial NaN values; custom filename for tx_file [#54]
• preprocessor.prep.JwstCalPrep updates in preparation for preprocessing spectroscopic data [#55]
  • revise spectroscopic data columns
  • save tx_file name with "-{expmode}" to differentiate between image and spec normalization params
  • rename target attributes: y_img_train, y_img_test to y_reg_train, y_reg_test
• preprocessor.scrub.JwstCalScrubber more sophisticated exposure grouping and L3 product naming [#56]

bug fixes

• preprocessor.encode.PairEncoder.handle_unknowns create single new encoding value per unidentified variable [#53]

installation / deps

• Install astropy dev from wheel by @pllim in #48
• Bump sphinx-automodapi from 0.15.0 to 0.17.0 by @dependabot in #60
• Bump plotly from 5.15.0 to 5.20.0 by @dependabot in #59
• Bump pydot from 1.4.2 to 2.0.0 by @dependabot in #62
• Bump pydot from 1.4.2 to 2.0.0 by @alphasentaurii in #64

New Contributors

• @pllim made their first contribution in #48

Full Changelog: 1.0.1...1.1.0

1.0.1

What's Changed

bugfixes by @alphasentaurii

• move HstSvmRadio import inside HstSvmScrubber class method to avoid importing astroquery unnecessarily #49
• matplotlib style setting looks for "seaborn-v0_8-bright" if "seaborn-bright" unavailable, fallback uses default style #49

installation / automation by @alphasentaurii

• temporarily pin tf max version to 2.15 to ensure compatibility with models saved in 2.13 or older #49
• GA workflow minor revision: pypi publish #46
• Replace flake8 with ruff, replace deprecated tf.keras.wrappers.scikit_learn with scikeras, add GA workflows #45

documentation by @alphasentaurii

• Update readthedocs.yaml for compatibility with latest formatting requirements [#44]
• RTD: Install graphviz before building docs [#47]

Full Changelog: 1.0.0...1.0.1

Spacekit v1.0.0

What's Changed

skopes/jwst/cal/predict by @alphasentaurii in #37

• added new skope for jwst calibration pipeline resource prediction modeling

jwst/minimal-install by @alphasentaurii in #38

• Default installation is a minimal set of dependencies. To install all dependencies, users must now use pip install spacekit[x]

compute/use-npz-instead-of-pickle by @alphasentaurii in #39

• Model training results files use compressed numpy files (.npz) and .csv instead of pickle.

preprocessor/skytransformer by @alphasentaurii in #40

• New class for estimating pixel offset calculations for a Level 3 image product using fiducial values of Level 1 exposures

pytest/compatibility-checks by @alphasentaurii in #41

• Compatibility updates for python 3.10 and 3.11 (backwards compatible with 3.9)

skopes/jwst-predict-tests by @alphasentaurii in #42

• New tests added for JWST skope and dependent scripts.
• Tests modified to reflect updates made to HST skopes and models being renamed

Full Changelog: 0.4.1...1.0.0

Spacekit v0.4.1

What's Changed

skopes/hst-cal-predict by @alphasentaurii in #34

• added predict script for hst cal skope
• enhancements for loading pretrained models
• updated docker dashboard templates
• improved log handling with spacekit/logger module

datasets/external-data by @alphasentaurii in #35

• plugin for external test data
• updated repo url badges
• updated documentation

pytest/cfg-updates-and-cal-s3-predict-tests by @alphasentaurii in #36

• bugfix set dataframe columns with bracket instead of curly bracket (resolves pandas>1.4 incompatibility)
• remove pandas pinned version
• pytest configuration updates and new tests added

Full Changelog: 0.4.0...0.4.1

Spacekit v0.4.0

What's Changed in 0.4.0 (2022-12-08)

• bugfix scikit-learn replaces deprecated sklearn dependency

• temporarily pinned pandas dependency to 1.4.x and below due to column setting bug in v1.5

• bugfix keras load_img method imported from tf.keras.preprocessing.image instead of tf.keras.utils

• new feature skopes.hst.cal model training, inference, cross-validation scripts added

• new feature svm dashboard predict view

• svm ensemble model archive file ensembleSVM.zip renamed as ensemble.zip. This extracts to models/ensemble/ with tx_data.json (transform data) and ensembleSVM (keras model binaries) inside of the ensemble/ parent directory. Previously, the json file was inside ensembleSVM alongside the binaries.

Full Changelog: alphasentaurii/spacekit@0.3.2...0.4.0

Spacekit v0.3.2

What's Changed

• Docker image deployment bugfixes and cleaner organization
• Updated calcloud model results formatting to conform with spacekit compute module I/O
• Bugfix for dataset scrape/import

Full Changelog: alphasentaurii/spacekit@0.3.1...0.3.2

Spacekit v0.3.1

• Bug fix relating to the SVM predict.py Classification Report which mistakenly assumed all categorical types are represented in the data (not necessarily the case for prediction inputs). Fixing the encoder resolves the issue (see below)

• A custom encoder class PairEncoder was created, allowing a user to pass in explicit key-pair values (a dictionary) for categorical features and SvmEncoder was updated to use this for encoding “category" (scene/field), "detector" and "wcs".

• Additional tests added to test_encode.py for the above case

• Minor enhancements to SVM classification report for better readability.

What's Changed

• Testing/0.3.0rc2 by @alphasentaurii in https://github.com/alphasentaurii/spacekit/pull/17
• fix/svm-pred-encoding by @alphasentaurii in https://github.com/alphasentaurii/spacekit/pull/20

Full Changelog: alphasentaurii/spacekit@0.3.0...0.3.1

Spacekit v0.3.0

-new feature SVM dashboard for model evaluation and data analysis
-enhancements to SVM prep, predict and training modules
-significant additions made to pytest test suite for primary svm-related modules
-minor bug fixes and enhancements
-ability to load/save image arrays as compressed numpy files (single .npz file instead of individual pngs).
-load dataset module added for calcloud dashboard
-Read the Docs documentation and API

What's Changed

• Release/0.3.0 #19
• docs/rtd-updates #11
• svm-testing by #12, #15, #16, #18
• svm/dashboard by #13, #14

Full Changelog: alphasentaurii/spacekit@0.2.8...0.3.0