Release v3.4.0

Release v3.4.0

Highlights

• New method to compute Kyoto-GHG from individual species by different GWP-metrics
• Improved support for yearly, datetime and categorical "subannual" timeseries data
• Integration with latest version of ixmp4 database package

API changes

• Change expected behavior of subannual column to use None for yearly data (instead of 'year')
• Rename the method aggregate_time() to aggregate_subannual()

v3.3.2

What's Changed

• Fix ixmp4 platform lookup via read_iiasa by @meksor in #972
• Adjust SceSeAuth to work with new authentication classes by @meksor in #974
• Add compatibility with latest ixmp4 by @danielhuppmann in #975
• Release candidate v3.3.2 by @danielhuppmann in #976

Full Changelog: v3.3.1...v3.3.2

v3.3.1

What's Changed

• Quick fixes following release v3.3.0 by @danielhuppmann in #963
• Quickfix for filtering by measurand as list with length > 2 by @danielhuppmann in #965
• Bump pygments from 2.19.2 to 2.20.0 by @dependabot[bot] in #964
• Fixes for ixmp4 v0.15.x by @meksor in #970

New Contributors

• @meksor made their first contribution in #970

Full Changelog: v3.3.0...v3.3.1

Release v3.3.0

Release v3.3.0

Highlights

• New method to compute shares of a timeseries indicator relative to a reference
• Improved integration with ixmp4
• Support for Python 3.14 and updated dependencies

Individual updates

• #961 Support Python 3.14 and bump dependencies
• #946 Handle plot-styling by meta indicators with np.nan
• #944 Refactor to a format_n() function for nice log messages
• #943 Improved handling for division by zero
• #941 Add a compute.share() method
• #938 Add a function to initialize an IamDataFrame from an ixmp4.Run
• #950 Add a useful error message on empty query result from an ixmp4.Platform

Release v3.2.0

Release v3.2.0

Highlights

This release fixes an issue in the categorize() method, adds a check against infinite values in timeseries data, and simplifies the behavior of binary operations (addition, subtraction, division) with non-SI units.

Individual updates

• #937 Check for infinite values timeseries
• #930 Fix the categorize() method if invalid datapoints exceed the number of scenarios
• #907 Add fast route for operations with non-SI units

Release v3.1.0

Release v3.1.0

Highlights

This release implements integration with the latest ixmp4 version including support
for datetime-format timeseries, versioning and database transactions.

Dependency changes

Bumped minimum version of ixmp4 to 0.13.0 for compatibility with versioning
and database transactions.

Release v3.0.0

Release v3.0.0

Highlights

• Implement reading and writing of netcdf files implemented by @LinhHo
• Remove automated sorting of timeseries data for performance improvement
• Add support Python 3.13

API changes

The internal object that holds the timeseries data is not automatically sorted any more, so the pd.DataFrame returned from IamDataFrame.data may change when upgrading to pyam v3.0. You can use the new method IamDataFrame.sort_data() to explicitly sort the timeseries data by the data dimensions.

Individual updates

• #901 Add support for Python 3.13
• #899 Add to_netcdf() method
• #896 Add sort_data() method
• #896 Sort columns of timeseries() with mixed time domain
• #893 No sorting of timeseries data on initialization or append
• #879 Add read_netcdf() function

New Contributors

• @LinhHo made their first contribution in #879

Full Changelog: v2.3.0...v3.0.0

Release v2.3.0

Release v2.3.0

Highlights

This release includes a new option to filter by variable depth.

Individual updates

• #886 Implement filtering by variable depth
• #880 Use pd.Series.iloc[pos] for forward-compatibility
• #877 Support engine and other pd.ExcelFile keywords

New Contributors

• @korsbakken made their first contribution in #877
• @quant12345 made their first contribution in #880

Release v2.2.4

Release v2.2.4

Support filtering by a measurand argument with tuples of variable and units and improve the filter() documentation.

Release v2.2.3

Release v2.2.3

Patch release to set minimum requirement of ixmp4 to latest version and pin numpy to <2.0.