Merge branch 'develop' into pre-commit-configuration

AUTHORS.md

@@ -29,8 +29,10 @@
 * Raphael Portmann
 * Nicolas Colombi
 * Leonie Villiger
+* Timo Schmid
 * Kam Lam Yeung
 * Sarah Hülsen
 * Timo Schmid
 * Luca Severino
 * Samuel Juhel
+* Valentin Gebhart

CHANGELOG.md

@@ -12,24 +12,73 @@ Code freeze date: YYYY-MM-DD
 
 ### Added
 
+### Changed
+
+### Fixed
+
+### Deprecated
+
+### Removed
+
+## 5.0.0
+
+Release date: 2024-07-19
+
+### Dependency Changes
+
+Added:
+
+- `bayesian-optimization`
+- `seaborn` >=0.13
+
+Updated:
+
+- `bottleneck` >=1.3 → >=1.4
+- `cartopy` >=0.22 → >=0.23
+- `contextily` >=1.5 → >=1.6
+- `dask` >=2024.1,<2024.3 &rarr; >=2024.2,<2024.3
+- `matplotlib-base` >=3.8 &rarr; >=3.9
+- `numba` >=0.59 &rarr; >=0.60
+- `numexpr` >=2.9 &rarr; >=2.10
+- `pint` >=0.23 &rarr; >=0.24
+- `pycountry` >=22.3 &rarr; >=24.6
+- `requests` >=2.31 &rarr; >=2.32
+- `salib` >=1.4 &rarr; >=1.5
+- `scikit-learn` >=1.4 &rarr; >=1.5
+- `scipy` >=1.12 &rarr; >=1.13
+- `xarray` >=2024.2 &rarr; >=2024.6
+
+### Added
+
 - GitHub actions workflow for CLIMADA Petals compatibility tests [#855](https://github.com/CLIMADA-project/climada_python/pull/855)
+- `climada.util.calibrate` module for calibrating impact functions [#692](https://github.com/CLIMADA-project/climada_python/pull/692)
 
 ### Changed
 
+- Use Geopandas GeoDataFrame.plot() for centroids plotting function [896](https://github.com/CLIMADA-project/climada_python/pull/896)
 - Update SALib sensitivity and sampling methods from newest version (SALib 1.4.7) [#828](https://github.com/CLIMADA-project/climada_python/issues/828)
 - Allow for computation of relative and absolute delta impacts in `CalcDeltaClimate`
 - Remove content tables and make minor improvements (fix typos and readability) in
 CLIMADA tutorials. [#872](https://github.com/CLIMADA-project/climada_python/pull/872)
 - Centroids complete overhaul. Most function should be backward compatible. Internal data is stored in a geodataframe attribute. Raster are now stored as points, and the meta attribute is removed. Several methds were deprecated or removed. [#787](https://github.com/CLIMADA-project/climada_python/pull/787)
 - Improved error messages produced by `ImpactCalc.impact()` in case impact function in the exposures is not found in impf_set [#863](https://github.com/CLIMADA-project/climada_python/pull/863)
+- Update the Holland et al. 2010 TC windfield model and introduce `model_kwargs` parameter to adjust model parameters [#846](https://github.com/CLIMADA-project/climada_python/pull/846)
 - Changed module structure: `climada.hazard.Hazard` has been split into the modules `base`, `io` and `plot` [#871](https://github.com/CLIMADA-project/climada_python/pull/871)
+- Ensure `csr_matrix` stored in `climada.hazard.Hazard` have consistent data format and store no explicit zeros when initializing `ImpactCalc` [#893](https://github.com/CLIMADA-project/climada_python/pull/893)
+- `Impact.from_hdf5` now calls `str` on `event_name` data that is not strings, and issue a warning then [#894](https://github.com/CLIMADA-project/climada_python/pull/894)
+- `Impact.write_hdf5` now throws an error if `event_name` is does not contain strings exclusively [#894](https://github.com/CLIMADA-project/climada_python/pull/894)
+- Split `climada.hazard.trop_cyclone` module into smaller submodules without affecting module usage [#911](https://github.com/CLIMADA-project/climada_python/pull/911)
 
 ### Fixed
 
 - Avoid an issue where a Hazard subselection would have a fraction matrix with only zeros as entries by throwing an error [#866](https://github.com/CLIMADA-project/climada_python/pull/866)
+- Allow downgrading the Python bugfix version to improve environment compatibility [#900](https://github.com/CLIMADA-project/climada_python/pull/900)
+- Fix broken links in `CONTRIBUTING.md` [#900](https://github.com/CLIMADA-project/climada_python/pull/900)
+- When writing `TCTracks` to NetCDF, only apply compression to `float` or `int` data types. This fixes a downstream issue, see [climada_petals#135](https://github.com/CLIMADA-project/climada_petals/issues/135) [#911](https://github.com/CLIMADA-project/climada_python/pull/911)
 
 ### Added
 
+- Method `Hazard.check_matrices` for bringing the stored CSR matrices into "canonical format" [#893](https://github.com/CLIMADA-project/climada_python/pull/893)
 - Generic s-shaped impact function via `ImpactFunc.from_poly_s_shape` [#878](https://github.com/CLIMADA-project/climada_python/pull/878)
 - climada.hazard.centroids.centr.Centroids.get_area_pixel
 - climada.hazard.centroids.centr.Centroids.get_dist_coast
@@ -50,6 +99,7 @@ CLIMADA tutorials. [#872](https://github.com/CLIMADA-project/climada_python/pull
 - climada.hazard.centroids.centr.Centroids.empty_geometry_points
 - climada.hazard.centroids.centr.Centroids.set_meta_to_lat_lon
 - climada.hazard.centroids.centr.Centroids.set_lat_lon_to_meta
+- `scheduler` parameter in `climada.util.coordinates.set_df_geometry_points`, as dask is not used anymore, leaving all calculation to shapely [#912](https://github.com/CLIMADA-project/climada_python/pull/912)
 
 ### Removed
 

CONTRIBUTING.md

@@ -22,7 +22,7 @@ Please contact the [lead developers](https://wcr.ethz.ch/research/climada.html)
 
 ## Minimal Steps to Contribute
 
-Before you start, please have a look at our [Developer Guide][devguide].
+Before you start, please have a look at our Developer Guide section in the [CLIMADA Docs][docs].
 
 To contribute follow these steps:
 
@@ -65,21 +65,22 @@ To contribute follow these steps:
 
 ## Resources
 
-The CLIMADA documentation provides a [Developer Guide][devguide].
+The [CLIMADA documentation][docs] provides several Developer Guides.
 Here's a selection of the commonly required information:
 
 * How to use Git and GitHub for CLIMADA development: [Development and Git and CLIMADA](https://climada-python.readthedocs.io/en/latest/guide/Guide_Git_Development.html)
-* Coding instructions for CLIMADA: [Python Dos and Don'ts](https://climada-python.readthedocs.io/en/latest/guide/Guide_PythonDos-n-Donts.html), [Performance Tips](https://climada-python.readthedocs.io/en/latest/guide/Guide_Py_Performance.html), [CLIMADA Conventions](https://climada-python.readthedocs.io/en/latest/guide/Guide_Miscellaneous.html)
-* How to execute tests in CLIMADA: [Testing and Continuous Integration][testing]
+* Coding instructions for CLIMADA: [Python Dos and Don'ts](https://climada-python.readthedocs.io/en/latest/guide/Guide_PythonDos-n-Donts.html), [Performance Tips](https://climada-python.readthedocs.io/en/latest/guide/Guide_Py_Performance.html), [CLIMADA Conventions](https://climada-python.readthedocs.io/en/latest/guide/Guide_CLIMADA_conventions.html)
+* How to execute tests in CLIMADA: [Testing][testing] and [Continuous Integration](https://climada-python.readthedocs.io/en/latest/guide/Guide_continuous_integration_GitHub_actions.html)
 
 ## Pull Requests
 
 After developing a new feature, fixing a bug, or updating the tutorials, you can create a [pull request](https://docs.github.com/en/pull-requests) to have your changes reviewed and then merged into the CLIMADA code base.
 To ensure that your pull request can be reviewed quickly and easily, please have a look at the _Resources_ above before opening a pull request.
-In particular, please check out the [Pull Request instructions](https://climada-python.readthedocs.io/en/latest/guide/Guide_Git_Development.html#Pull-requests).
+In particular, please check out the [Pull Request instructions](https://climada-python.readthedocs.io/en/latest/guide/Guide_Git_Development.html#pull-requests).
 
 We provide a description template for pull requests that helps you provide the essential information for reviewers.
 It also contains a checklist for both pull request authors and reviewers to guide the review process.
 
+[docs]: https://climada-python.readthedocs.io/en/latest/
 [devguide]: https://climada-python.readthedocs.io/en/latest/#developer-guide
-[testing]: https://climada-python.readthedocs.io/en/latest/guide/Guide_Continuous_Integration_and_Testing.html
+[testing]: https://climada-python.readthedocs.io/en/latest/guide/Guide_Testing.html

climada/_version.py

@@ -1 +1 @@
-__version__ = '4.1.2-dev'
+__version__ = '5.0.1-dev'

climada/conf/climada.conf

@@ -67,7 +67,7 @@
         "cache_db": "{local_data.system}/.downloads.db",
         "cache_enabled": true,
         "cache_dir": "{local_data.system}/.apicache",
-        "supported_hazard_types": ["river_flood", "tropical_cyclone", "storm_europe", "relative_cropyield", "wildfire", "earthquake", "flood", "hail"],
+        "supported_hazard_types": ["river_flood", "tropical_cyclone", "storm_europe", "relative_cropyield", "wildfire", "earthquake", "flood", "hail", "aqueduct_coastal_flood"],
         "supported_exposures_types": ["litpop", "crop_production", "ssp_population", "crops"]
     }
 }

climada/engine/impact.py

@@ -937,11 +937,6 @@ def write_hdf5(self, file_path: Union[str, Path], dense_imp_mat: bool=False):
 
         The impact matrix can be stored in a sparse or dense format.
 
-        Notes
-        -----
-        This writer does not support attributes with variable types. Please make sure
-        that ``event_name`` is a list of equally-typed values, e.g., all ``str``.
-
         Parameters
         ----------
         file_path : str or Path
@@ -950,6 +945,11 @@ def write_hdf5(self, file_path: Union[str, Path], dense_imp_mat: bool=False):
             If ``True``, write the impact matrix as dense matrix that can be more easily
             interpreted by common H5 file readers but takes up (vastly) more space.
             Defaults to ``False``.
+
+        Raises
+        ------
+        TypeError
+            If :py:attr:`event_name` does not contain strings exclusively.
         """
         # Define writers for all types (will be filled later)
         type_writers = dict()
@@ -983,7 +983,7 @@ def write(group: h5py.Group, name: str, value: Any):
 
         def _str_type_helper(values: Collection):
             """Return string datatype if we assume 'values' contains strings"""
-            if isinstance(next(iter(values)), str):
+            if all((isinstance(val, str) for val in values)):
                 return h5py.string_dtype()
             return None
 
@@ -1037,6 +1037,8 @@ def write_csr(group, name, value):
             # Now write all attributes
             # NOTE: Remove leading underscore to write '_tot_value' as regular attribute
             for name, value in self.__dict__.items():
+                if name == "event_name" and _str_type_helper(value) is None:
+                    raise TypeError("'event_name' must be a list of strings")
                 write(file, name.lstrip("_"), value)
 
     def write_sparse_csr(self, file_name):
@@ -1240,10 +1242,18 @@ def from_hdf5(cls, file_path: Union[str, Path]):
             ).intersection(file.keys())
             kwargs.update({attr: file[attr][:] for attr in array_attrs})
 
-            # Special handling for 'event_name' because it's a list of strings
+            # Special handling for 'event_name' because it should be a list of strings
             if "event_name" in file:
                 # pylint: disable=no-member
-                kwargs["event_name"] = list(file["event_name"].asstr()[:])
+                try:
+                    event_name = file["event_name"].asstr()[:]
+                except TypeError:
+                    LOGGER.warning(
+                        "'event_name' is not stored as strings. Trying to decode "
+                        "values with 'str()' instead."
+                    )
+                    event_name = map(str, file["event_name"][:])
+                kwargs["event_name"] = list(event_name)
 
         # Create the impact object
         return cls(**kwargs)

climada/engine/impact_calc.py

@@ -47,6 +47,8 @@ def __init__(self,
         The dimension of the imp_mat variable must be compatible with the
         exposures and hazard objects.
 
+        This will call :py:meth:`climada.hazard.base.Hazard.check_matrices`.
+
         Parameters
         ----------
         exposures : climada.entity.Exposures
@@ -61,6 +63,8 @@ def __init__(self,
         self.exposures = exposures
         self.impfset = impfset
         self.hazard = hazard
+        self.hazard.check_matrices()
+
         # exposures index to use for matrix reconstruction
         self._orig_exp_idx = np.arange(self.exposures.gdf.shape[0])
 

climada/engine/test/test_impact.py

@@ -779,7 +779,8 @@ def test_select_event_identity_pass(self):
         ent.exposures.assign_centroids(hazard)
 
         # Compute the impact over the whole exposures
-        imp = ImpactCalc(ent.exposures, ent.impact_funcs, hazard).impact(save_mat=True, assign_centroids=False)
+        imp = ImpactCalc(ent.exposures, ent.impact_funcs, hazard).impact(
+            save_mat=True, assign_centroids=False)
 
         sel_imp = imp.select(event_ids=imp.event_id,
                              event_names=imp.event_name,
@@ -1019,10 +1020,11 @@ def test_write_hdf5_without_imp_mat(self):
 
     def test_write_hdf5_type_fail(self):
         """Test that writing attributes with varying types results in an error"""
-        self.impact.event_name = [1, "a", 1.0, "b", "c", "d"]
-        with self.assertRaises(TypeError) as cm:
+        self.impact.event_name = ["a", 1, 1.0, "b", "c", "d"]
+        with self.assertRaisesRegex(
+            TypeError, "'event_name' must be a list of strings"
+        ):
             self.impact.write_hdf5(self.filepath)
-        self.assertIn("No conversion path for dtype", str(cm.exception))
 
     def test_cycle_hdf5(self):
         """Test writing and reading the same object"""
@@ -1120,6 +1122,15 @@ def test_read_hdf5_full(self):
         impact = Impact.from_hdf5(self.filepath)
         npt.assert_array_equal(impact.imp_mat.toarray(), [[0, 1, 2], [3, 0, 0]])
 
+        # Check with non-string event_name
+        event_name = [1.2, 2]
+        with h5py.File(self.filepath, "r+") as file:
+            del file["event_name"]
+            file.create_dataset("event_name", data=event_name)
+        with self.assertLogs("climada.engine.impact", "WARNING") as cm:
+            impact = Impact.from_hdf5(self.filepath)
+        self.assertIn("'event_name' is not stored as strings", cm.output[0])
+        self.assertListEqual(impact.event_name, ["1.2", "2.0"])
 
 # Execute Tests
 if __name__ == "__main__":

climada/engine/test/test_impact_calc.py

@@ -70,6 +70,16 @@ def test_init(self):
         np.testing.assert_array_equal(HAZ.event_id, icalc.hazard.event_id)
         np.testing.assert_array_equal(HAZ.event_name, icalc.hazard.event_name)
 
+        # Test check matrices
+        hazard = deepcopy(HAZ)
+        hazard.intensity[0, hazard.intensity.indices[0]] = 0
+        hazard.fraction = sparse.csr_matrix(np.ones((1, 1)))
+        with self.assertRaisesRegex(
+            ValueError, "Intensity and fraction matrices must have the same shape"
+        ):
+            ImpactCalc(ENT.exposures, ENT.impact_funcs, hazard)
+            self.assertEqual(hazard.intensity.nnz, HAZ.intensity.nnz - 1)  # was pruned
+
     def test_metrics(self):
         """Test methods to get impact metrics"""
         mat = sparse.csr_matrix(np.array(