Changelog

v0.9.0 - unreleased

New Features

• Extracted statistical functionality for linear regression:

  • Create mesmer.stats.LinearRegression which encapsulates fit, predict, etc. methods around linear regression (#134). By Mathias Hauser.

  • Add mesmer.stats._fit_linear_regression_xr: xarray wrapper for mesmer.stats._fit_linear_regression_np (#123 and #142). By Mathias Hauser.

  • Add add fit_intercept argument to the linear_regression fitting methods and functions (#144). By Mathias Hauser.

  • Allow to pass 1-dimensional targets to mesmer.stats.LinearRegression.fit() (#221). By Mathias Hauser.

  • Fixed two bugs related to (non-dimension) coordinates ( #332, #333 and #334). By Mathias Hauser.

• Extracted statistical functionality for auto regression:

  • Add mesmer.stats.fit_auto_regression: xarray wrapper to fit an auto regression model (#139). By Mathias Hauser.

  • Have mesmer.stats.fit_auto_regression return the variance instead of the standard deviation ( #306 #318). By Mathias Hauser.

  • Add draw_auto_regression_correlated and draw_auto_regression_uncorrelated: to draw samples of a (spatially-)correlated and uncorrelated auto regression model ( #322, #161 and #313). By Mathias Hauser.

  • Add mesmer.stats.select_ar_order to select the order of an auto regressive model (#176). By Mathias Hauser.

• Extracted functions dealing with the spatial covariance and its localization:

  • Add xarray wrappers mesmer.stats.adjust_covariance_ar1() and mesmer.stats.find_localized_empirical_covariance() (#191). By Mathias Hauser.

  • Refactor and extract numpy-based functions dealing with the spatial covariance and its localization (#167 and #184). By Mathias Hauser.

  • Allow to pass 1 x n arrays to mesmer.stats.adjust_covariance_ar1() (#224). By Mathias Hauser.

• Update LOWESS smoothing:

  • Extract the LOWESS smoothing for xarray objects: mesmer.stats.lowess(). (#193, #283, and #285). By Mathias Hauser.

  • Allow to pool data along a dimension to estimate the LOWESS smoothing. (#331). By Mathias Hauser.

• Added helper functions to process xarray-based model data:

  • Added functions to stack regular lat-lon grids to 1D grids and unstack them again (#217). By Mathias Hauser.

  • Added functions to mask the ocean and Antarctica ( #219 and #314). By Mathias Hauser.

  • Added functions to calculate the weighted global mean (#220 and #287). By Mathias Hauser.

  • Added functions to wrap arrays to [-180, 180) and [0, 360), respectively (#270 and #273). By Mathias Hauser.

• The aerosol data is now automatically downloaded using pooch. (#267). By Mathias Hauser.

• Added helper functions to estimate and superimpose volcanic influence (#336). By Mathias Hauser.

• Added additional tests for the calibration step (#209):

  • one scenario (SSP5-8.5) and two ensemble members (#211)

  • two scenarios (SSP1-2.6 and SSP5-8.5) with one and two ensemble members, respectively (#214)

  • different selection of predictor variables (tas**2 and hfds) for different scenarios (#291)

  By Mathias Hauser.

• Allow passing xr.DataArray to gaspari_cohn (#298). By Mathias Hauser.

• Allow passing xr.DataArray to geodist_exact (#299). By Zeb Nicholls and Mathias Hauser.

• Add calc_gaspari_cohn_correlation_matrices a function to calculate Gaspari-Cohn correlation matrices for a range of localisation radii (#300). By Zeb Nicholls and Mathias Hauser.

• Add a helper function to load tas and (potentially) hfds for several ESMs from cmip-ng archive at ETHZ (#326). By Mathias Hauser.

Breaking changes

• Localization radii that lead to singular matrices are now skipped (#187). By Mathias Hauser.

• Refactor and split train_l_prepare_X_y_wgteq() into two functions: get_scenario_weights() and stack_predictors_and_targets() (#143). By Mathias Hauser.

• Moved gaspari_cohn & calc_geodist_exact from io.load_constant_files to core.computation (#158). By Yann Quilcaille.

• The function mask_percentage has been renamed to core.regionmaskcompat.mask_3D_frac_approx() (#202). By Mathias Hauser.

• Removed mesmer.io.load_constant_files.infer_interval_breaks() and the edges from the lat and lon dictionaries i.e., lon["e"] and lat["e"] (#233). By Mathias Hauser.

• Deprecated the reg_type argument to mesmer.io.load_constant_files.load_regs_ls_wgt_lon_lat() and the reg_dict argument to mesmer.utils.select.extract_land(). These arguments no longer have any affect (#235). By Mathias Hauser.

• Removed ref["type"] == "first", i.e., calculating the anomaly w.r.t. the first ensemble member (#247). By Mathias Hauser.

• Renamed mesmer.calibrate_mesmer._calibrate_and_draw_realisations to mesmer.calibrate_mesmer._calibrate_tas (#66). By Mathias Hauser.

Deprecations

• The function mesmer.create_emulations.create_emus_gt has been renamed to create_emulations.gather_gt_data() (#246). By Mathias Hauser.

• The function mesmer.utils.select.extract_time_period is now deprecated and will be removed in a future version. Please raise an issue if you use this function (#243). By Mathias Hauser.

Bug fixes

• Fix three issues with core.regionmaskcompat.mask_3D_frac_approx(). Note that these issues are only relevant if passing xarray objects and/ or masks close to the poles (#202 and #218). By Mathias Hauser.

Documentation

• Add development/contributing docs (#121). By Zeb Nicholls.

Internal Changes

• Refactor the mesmer internals to use the new statistical core, employ helper functions etc.:

  • Use mesmer.utils.separate_hist_future() in mesmer.calibrate_mesmer.train_gt() (#281).

  • Use of mesmer.stats.LinearRegression in

    • mesmer.calibrate_mesmer.train_gt_ic_OLSVOLC() (#145).

    • mesmer.create_emulations.create_emus_lv_OLS() and mesmer.create_emulations.create_emus_OLS_each_gp_sep() (#240).

  By Mathias Hauser.

• Restore compatibility with regionmask v0.9.0 (#136). By Mathias Hauser.

• Renamed the interpolation keyword of np.quantile to method changed in numpy v1.22.0 (#137). By Mathias Hauser.

• Add python 3.10 and python 3.11 to list of supported versions (#162 and #284). By Mathias Hauser.

• Move contents of setup.py to setup.cfg (#169). By Mathias Hauser.

• Use pyproject.toml for the build-system and setuptools_scm for the __version__ (#188). By Mathias Hauser.

• Moved the climate model data manipulation functions (#237). By Mathias Hauser.

v0.8.3 - 2021-12-23

New Features

• Add mesmer.stats._linear_regression (renamed to mesmer.stats._fit_linear_regression_np in #142). Starts the process of refactoring the codebase (#116). By Zeb Nicholls.

Bug fixes

• Initialize llh_max to -inf to ensure the cross validation loop is entered (#110). By Jonas Schwaab.

Documentation

• Fix copyright notice and release version in documentation (#127). By Zeb Nicholls.

Internal Changes

• Automatically upload the code coverage to codecov.io after the test suite has run (#99). By Mathias Hauser.

• Internal refactor: moved a number of inline comments to their own line (especially if this allows to have the code on one line instead of several) and other minor cleanups (#98). By Mathias Hauser.

• Refactor io.load_cmipng_tas and io.load_cmipng_hfds to de-duplicate their code and add tests for them (#55). By Mathias Hauser.

v0.8.2 - 2021-10-07

Bug fixes

• Reintroduce ability to read in cmip5 data from the cmip5-ng archive at ETH (#90). By Lea Beusch.

Internal Changes

• Reproduce the test files because of a change in regionmask which affected the mesmer tests (#95). By Mathias Hauser.

• Refactor and speed up of the Gaspari-Cohn function and the calculation of the great circle distance (#85, #88). By Mathias Hauser.

• The geopy package is no longer a dependency of mesmer (#88). By Mathias Hauser.

• Convert README from Markdown to reStructuredText to fix package build errors. Also allows to include the README in the docs to avoid duplication (#102). By Mathias Hauser.

v0.8.1 - 2021-07-15

• Update example script (#80).

v0.8.0 - 2021-07-13

• First release on PyPI and conda (#79).