Merge pull request #799 from cgre-aachen/gempy_v2023.1.0-pandas2.0

[WIP] remove inplace for pandas cat
gempy-project · Jun 2, 2023 · 28dcdc1 · 28dcdc1
2 parents 3314aed + 2e0b2cd
commit 28dcdc1
Show file tree

Hide file tree

Showing 28 changed files with 145 additions and 98 deletions.
diff --git a/.github/workflows/pytest.yml b/.github/workflows/pytest.yml
@@ -60,7 +60,7 @@ jobs:
 
       - name: Run tests with pytest
         run: |
-          pytest
+          pytest --ignore-glob=/home/runner/work/gempy/gempy/gempy_3/api/test*_gempy3.py
 
   testing_pip:
     name: Pip install ${{ matrix.case.os }} py${{ matrix.case.python-version }} ${{ matrix.case.name }}
@@ -108,4 +108,4 @@ jobs:
 
       - name: Run tests with pytest
         run: |
-          pytest
+          pytest --ignore-glob=/home/runner/work/gempy/gempy/gempy_3/api/test*_gempy3.py
diff --git a/README.md b/README.md
@@ -11,11 +11,6 @@
 [![DOI](https://zenodo.org/badge/96211155.svg)](https://zenodo.org/badge/latestdoi/96211155)
 [![DOCKER](https://img.shields.io/docker/cloud/automated/leguark/gempy.svg)](https://cloud.docker.com/repository/docker/leguark/gempy)
 
-:warning: **Warning: GemPy requires pandas version < 1.4.0. The new pandas release is not compatible with GemPy.  
-    We're actively working on this issue for a future release.  
-Please make sure to use Pandas version 1.3.x when working with GemPy for the time being.** 
-
-**Using theano, GemPy requires numpy version < 1.22.0 as `blas_opt_info` was deprecated in newer numpy versions.**:warning:
 ## Overview
 
 [GemPy](https://www.gempy.org/) is a Python-based, **open-source geomodeling library**. It is
@@ -40,9 +35,17 @@ the aesara conda installation. Therefore the process would be the following:
 
 For more information, refer to the [installation documentation](https://docs.gempy.org/installation.html).
 
+## Requirements
+
+The following versions are required/strongly recommended for the main dependencies of GemPy (as of June 2023):
+- python>=3.10
+- pandas>=2.0
+- matplotlib>=3.7
+- pyvista>=0.39
+
 ## Resources
 
-After installation you can either check the [notebook tutorials](https://docs.gempy.org/getting_started/get_started.html#sphx-glr-getting-started-get-started-py) 
+After installation, you can either check the [notebook tutorials](https://docs.gempy.org/getting_started/get_started.html#sphx-glr-getting-started-get-started-py) 
 or the [video introduction](https://www.youtube.com/watch?v=n0btC5Zilyc) to get started.
 
 Go to the [documentation site](http://docs.gempy.org/) for further information and enjoy the [tutorials and examples](https://www.gempy.org/tutorials).
@@ -58,8 +61,8 @@ Follow these [guidelines](https://github.com/cgre-aachen/gempy/blob/WIP_readme-u
 
 * de la Varga, M., Schaaf, A., and Wellmann, F. (2019). [GemPy 1.0: open-source stochastic geological modeling and inversion](https://gmd.copernicus.org/articles/12/1/2019/gmd-12-1-2019.pdf), Geosci. Model Dev., 12, 1-32.
 * Wellmann, F., & Caumon, G. (2018). [3-D Structural geological models: Concepts, methods, and uncertainties.](https://hal.univ-lorraine.fr/hal-01921494/file/structural_models_for_geophysicsHAL.pdf) In Advances in Geophysics (Vol. 59, pp. 1-121). Elsevier.
-* Calcagno, P., Chilès, J. P., Courrioux, G., & Guillen, A. (2008). Geological modelling from field data and geological knowledge: Part I. Modelling method coupling 3D potential-field interpolation and geological rules. Physics of the Earth and Planetary Interiors, 171(1-4), 147-157.
-* Lajaunie, C., Courrioux, G., & Manuel, L. (1997). Foliation fields and 3D cartography in geology: principles of a method based on potential interpolation. Mathematical Geology, 29(4), 571-584.
+* Calcagno, P., Chilès, J. P., Courrioux, G., & Guillen, A. (2008). [Geological modelling from field data and geological knowledge: Part I. Modelling method coupling 3D potential-field interpolation and geological rules](https://www.sciencedirect.com/science/article/abs/pii/S0031920108001258). Physics of the Earth and Planetary Interiors, 171(1-4), 147-157.
+* Lajaunie, C., Courrioux, G., & Manuel, L. (1997). [Foliation fields and 3D cartography in geology: principles of a method based on potential interpolation](https://link.springer.com/article/10.1007/BF02775087). Mathematical Geology, 29(4), 571-584.
 
 ## Publications using GemPy
 

diff --git a/gempy/api_modules/io.py b/gempy/api_modules/io.py
@@ -157,10 +157,8 @@ def _load_surface_points(cat_series, cat_surfaces, geo_model, name, path):
                                                       'series': 'category',
                                                       'id': 'int64',
                                                       'order_series': 'int64'})
-    geo_model._surface_points.df['surface'].cat.set_categories(cat_surfaces,
-                                                               inplace=True)
-    geo_model._surface_points.df['series'].cat.set_categories(cat_series,
-                                                              inplace=True)
+    geo_model._surface_points.df['surface'] = geo_model._surface_points.df['surface'].cat.set_categories(cat_surfaces)
+    geo_model._surface_points.df['series'] = geo_model._surface_points.df['series'].cat.set_categories(cat_series)
     # Code to add smooth columns for models saved before gempy 2.0bdev4
     try:
         geo_model._surface_points.df['smooth']
@@ -183,9 +181,8 @@ def _load_orientations(cat_series, cat_surfaces, geo_model, name, path):
                                                     'series': 'category',
                                                     'id': 'int64',
                                                     'order_series': 'int64'})
-    geo_model._orientations.df['surface'].cat.set_categories(cat_surfaces,
-                                                             inplace=True)
-    geo_model._orientations.df['series'].cat.set_categories(cat_series, inplace=True)
+    geo_model._orientations.df['surface'] = geo_model._orientations.df['surface'].cat.set_categories(cat_surfaces)
+    geo_model._orientations.df['series'] = geo_model._orientations.df['series'].cat.set_categories(cat_series)
 
     try:
         geo_model._orientations.df['smooth']
@@ -207,7 +204,7 @@ def _load_surfaces(cat_series, geo_model, name, path):
         ordered=False)
     geo_model._surfaces.sort_surfaces()
     geo_model._surfaces.colors.generate_colordict()
-    geo_model._surfaces.df['series'].cat.set_categories(cat_series, inplace=True)
+    geo_model._surfaces.df['series'] = geo_model._surfaces.df['series'].cat.set_categories(cat_series)
     try:
         geo_model._surfaces.df['isActive']
     except KeyError:
@@ -228,8 +225,8 @@ def _load_stack(geo_model, name, path):
     series_index = pn.CategoricalIndex(geo_model._stack.df.index.values)
     # geo_model.series.df.index = pn.CategoricalIndex(series_index)
     geo_model._stack.df.index = series_index
-    geo_model._stack.df['BottomRelation'].cat.set_categories(
-        ['Erosion', 'Onlap', 'Fault'], inplace=True)
+    geo_model._stack.df['BottomRelation'] = geo_model._stack.df['BottomRelation'].cat.set_categories(
+        ['Erosion', 'Onlap', 'Fault'])
     try:
         geo_model._stack.df['isActive']
     except KeyError:
@@ -259,15 +256,14 @@ def _load_additional_data(geo_model, name, path):
                                                                'aesara_optimizer': 'category',
                                                                'device': 'category',
                                                                'verbosity': object})
-    geo_model._additional_data.options.df['dtype'].cat.set_categories(
-        ['float32', 'float64'], inplace=True)
-    geo_model._additional_data.options.df['aesara_optimizer'].cat.set_categories(
-        ['fast_run', 'fast_compile'],
-        inplace=True)
-    geo_model._additional_data.options.df['device'].cat.set_categories(
-        ['cpu', 'cuda'], inplace=True)
-    geo_model._additional_data.options.df['output'].cat.set_categories(
-        ['geology', 'gradients'], inplace=True)
+    geo_model._additional_data.options.df['dtype'] = geo_model._additional_data.options.df['dtype'].cat.set_categories(
+        ['float32', 'float64'])
+    geo_model._additional_data.options.df['aesara_optimizer'] = geo_model._additional_data.options.df['aesara_optimizer'].cat.set_categories(
+        ['fast_run', 'fast_compile'])
+    geo_model._additional_data.options.df['device'] = geo_model._additional_data.options.df['device'].cat.set_categories(
+        ['cpu', 'cuda'])
+    geo_model._additional_data.options.df['output'] = geo_model._additional_data.options.df['output'].cat.set_categories(
+        ['geology', 'gradients'])
     geo_model._additional_data.options.df.loc['values', 'verbosity'] = None
 
 

diff --git a/gempy/core/data.py b/gempy/core/data.py
@@ -31,9 +31,9 @@ def __init__(self):
                               'aesara_optimizer': 'category', 'device': 'category',
                               'verbosity': object})
 
-        self.df['dtype'].cat.set_categories(['float32', 'float64'], inplace=True)
-        self.df['aesara_optimizer'].cat.set_categories(['fast_run', 'fast_compile'], inplace=True)
-        self.df['device'].cat.set_categories(['cpu', 'cuda'], inplace=True)
+        self.df['dtype'] = self.df['dtype'].cat.set_categories(['float32', 'float64'])
+        self.df['aesara_optimizer'] = self.df['aesara_optimizer'].cat.set_categories(['fast_run', 'fast_compile'])
+        self.df['device'] = self.df['device'].cat.set_categories(['cpu', 'cuda'])
 
         self.default_options()
 

diff --git a/gempy/core/data_modules/geometric_data.py b/gempy/core/data_modules/geometric_data.py
@@ -42,7 +42,7 @@ def init_dependent_properties(self):
         self.df['series'] = 'Default series'
         self.df['series'] = self.df['series'].astype('category', copy=True)
 
-        self.df['series'].cat.set_categories(self.surfaces.df['series'].cat.categories, inplace=True)
+        self.df['series'] = self.df['series'].cat.set_categories(self.surfaces.df['series'].cat.categories)
 
         # id
         self.df['id'] = np.nan
@@ -80,12 +80,12 @@ def set_series_categories_from_series(self, series):
         Args:
             series (:class:`Series`): [s0]
         """
-        self.df['series'].cat.set_categories(series.df.index, inplace=True)
+        self.df['series'] = self.df['series'].cat.set_categories(series.df.index)
         return True
 
     def update_series_category(self):
         """Update the series categorical columns with the series categories of the :class:`Surfaces` attribute."""
-        self.df['series'].cat.set_categories(self.surfaces.df['series'].cat.categories, inplace=True)
+        self.df['series'] = self.df['series'].cat.set_categories(self.surfaces.df['series'].cat.categories)
 
         return True
 
@@ -98,7 +98,7 @@ def set_surface_categories_from_surfaces(self, surfaces: Surfaces):
 
         """
 
-        self.df['surface'].cat.set_categories(surfaces.df['surface'], inplace=True)
+        self.df['surface'] = self.df['surface'].cat.set_categories(surfaces.df['surface'])
         return True
 
     # @_setdoc_pro(Series.__doc__)
@@ -126,7 +126,7 @@ def map_data_from_series(self, series, attribute: str, idx=None):
 
         if type(self.df['order_series'].dtype) is pn.CategoricalDtype:
 
-            self.df['order_series'].cat.remove_unused_categories(inplace=True)
+            self.df['order_series'] = self.df['order_series'].cat.remove_unused_categories()
         return self
 
     @_setdoc_pro(Surfaces.__doc__)
@@ -165,7 +165,7 @@ def _add_surface_to_list_from_new_surface_points_or_orientations(self, idx, surf
         # Check is self.df['surface'] is a category
         if not isinstance(self.df['surface'].dtype, pd.CategoricalDtype):
             self.df['surface'] = self.df['surface'].astype('category', copy=True)
-            self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values, inplace=True)
+            self.df['surface'] = self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values)
 
         # Check if elements in surface are categories in self.df['surface'] and if not add them
         # for s in surface:

diff --git a/gempy/core/data_modules/orientations.py b/gempy/core/data_modules/orientations.py
@@ -66,7 +66,7 @@ def set_orientations(self, coord: np.ndarray = None, pole_vector: np.ndarray = N
                                         'azimuth', 'polarity', 'surface'], dtype=float)
 
         self.df['surface'] = self.df['surface'].astype('category', copy=True)
-        self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values, inplace=True)
+        self.df['surface'] = self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values)
 
         pole_vector = check_for_nans(pole_vector)
         orientation = check_for_nans(orientation)
@@ -94,7 +94,7 @@ def set_orientations(self, coord: np.ndarray = None, pole_vector: np.ndarray = N
                                          'this point. Check previous condition')
 
         self.df['surface'] = self.df['surface'].astype('category', copy=True)
-        self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values, inplace=True)
+        self.df['surface'] = self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values)
 
         self.init_dependent_properties()
 
@@ -161,10 +161,10 @@ def add_orientation(self, x, y, z, surface: list[str] | str, pole_vector: Union[
 
         # create new pandas categories from slef.df.['surface']
         self.df['surface'] = self.df['surface'].astype('category', copy=True)
-        self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values, inplace=True)
+        self.df['surface'] = self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values)
 
         self.df['series'] = self.df['series'].astype('category', copy=True)
-        self.df['series'].cat.set_categories(self.surfaces.df['series'].cat.categories, inplace=True)
+        self.df['series'] = self.df['series'].cat.set_categories(self.surfaces.df['series'].cat.categories)
 
         self.map_data_from_surfaces(self.surfaces, 'series', idx=idx)
         self.map_data_from_surfaces(self.surfaces, 'id', idx=idx)

diff --git a/gempy/core/data_modules/surface_points.py b/gempy/core/data_modules/surface_points.py
@@ -63,7 +63,7 @@ def set_surface_points(self, coord: np.ndarray = None, surface: list = None):
             self.df['surface'] = surface
 
         self.df['surface'] = self.df['surface'].astype('category', copy=True)
-        self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values, inplace=True)
+        self.df['surface'] = self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values)
 
         # Choose types
         self.init_dependent_properties()
@@ -128,10 +128,10 @@ def add_surface_points(self, x: Union[float, np.ndarray], y: Union[float, np.nda
         self.df.loc[idx, ['smooth']] = 1e-6
 
         self.df['surface'] = self.df['surface'].astype('category', copy=True)
-        self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values, inplace=True)
+        self.df['surface'] = self.df['surface'].cat.set_categories(self.surfaces.df['surface'].values)
 
         self.df['series'] = self.df['series'].astype('category', copy=True)
-        self.df['series'].cat.set_categories(self.surfaces.df['series'].cat.categories, inplace=True)
+        self.df['series'] = self.df['series'].cat.set_categories(self.surfaces.df['series'].cat.categories)
 
         self.map_data_from_surfaces(self.surfaces, 'series', idx=idx)
         self.map_data_from_surfaces(self.surfaces, 'id', idx=idx)

diff --git a/gempy/core/model.py b/gempy/core/model.py
@@ -621,9 +621,8 @@ def modify_order_features(self, new_value: int, idx: str):
         """
         self._stack.modify_order_series(new_value, idx)
 
-        self._surfaces.df['series'].cat.reorder_categories(
-            np.asarray(self._stack.df.index),
-            ordered=False, inplace=True)
+        self._surfaces.df['series'] = self._surfaces.df['series'].cat.reorder_categories(
+            np.asarray(self._stack.df.index), ordered=False)
 
         self._surfaces.sort_surfaces()
         self._surfaces.set_basement()

diff --git a/gempy/core/surfaces.py b/gempy/core/surfaces.py
@@ -48,7 +48,7 @@ def __init__(self, series, surface_names=None, values_array=None, properties_nam
         if (np.array(sys.version_info[:2]) <= np.array([3, 6])).all():
             self.df: pn.DataFrame
 
-        self.df['series'].cat.add_categories(['Default series'], inplace=True)
+        self.df['series'] = self.df['series'].cat.add_categories(['Default series'])
         if surface_names is not None:
             self.set_surfaces_names(surface_names)
 
@@ -64,7 +64,7 @@ def __repr__(self):
     def _repr_html_(self):
         c_ = self.df.columns[~(self.df.columns.isin(self._columns_vis_drop))]
 
-        return self.df[c_].style.applymap(self.background_color, subset=['color']).render()
+        return self.df[c_].style.applymap(self.background_color, subset=['color']).to_html()
 
     @property
     def properties_val(self):
@@ -312,7 +312,7 @@ def map_series(self, mapping_object: Union[dict, pn.DataFrame] = None):
         """
 
         # Updating surfaces['series'] categories
-        self.df['series'].cat.set_categories(self.series.df.index, inplace=True)
+        self.df['series'] = self.df['series'].cat.set_categories(self.series.df.index)
         # TODO Fixing this. It is overriding the formations already mapped
         if mapping_object is not None:
             # If none is passed and series exists we will take the name of the first series as a default
@@ -326,8 +326,8 @@ def map_series(self, mapping_object: Union[dict, pn.DataFrame] = None):
                         s.append(k)
                         f.append(form)
 
-                new_series_mapping = pn.DataFrame([pn.Categorical(s, self.series.df.index)],
-                                                  f, columns=['series'])
+                new_series_mapping = pn.DataFrame(list(zip(pn.Categorical(s, self.series.df.index))),
+                                                 index=f, columns=['series'])
 
             elif isinstance(mapping_object, pn.Categorical):
                 # This condition is for the case we have surface on the index and in 'series' the category
@@ -351,6 +351,7 @@ def map_series(self, mapping_object: Union[dict, pn.DataFrame] = None):
         self.reset_order_surfaces()
         self.sort_surfaces()
         self.set_basement()
+
         return self
 
     # endregion

diff --git a/..._api/test_model_gempy2_to_gempy3_basic.py → ..._api/test_basic_model_gempy2_to_gempy3.py b/..._api/test_model_gempy2_to_gempy3_basic.py → ..._api/test_basic_model_gempy2_to_gempy3.py
@@ -1,24 +1,31 @@
 import os
+import pytest
+
+#IN_GITHUB_ACTIONS = os.getenv("GITHUB_ACTIONS") == "true"
+IN_GITHUB_ACTIONS = True
+
 import gempy as gp
-import gempy_engine
 from gempy import Project
 from gempy.plot.vista import GemPyToVista
 from gempy_3.api.gp2_to_gp3_input import gempy_project_to_interpolation_input, gempy_project_to_input_data_descriptor, gempy_project_to_interpolation_options
 from gempy_3.api.gp3_to_gp2_output import set_gp3_solutions_to_gp2_solution
 from gempy_3.api.test_api._gp2togp3_test_utils import create_interpolator, load_model, map_sequential_pile
-from gempy_engine.config import AvailableBackends
-from gempy_engine.core.backend_tensor import BackendTensor
-from gempy_engine.core.data import InterpolationOptions
-from gempy_engine.core.data.input_data_descriptor import InputDataDescriptor
-from gempy_engine.core.data.interpolation_input import InterpolationInput
-from gempy_engine.core.data.solutions import Solutions
+
+if not IN_GITHUB_ACTIONS:
+    import gempy_engine
+    from gempy_engine.config import AvailableBackends
+    from gempy_engine.core.backend_tensor import BackendTensor
+    from gempy_engine.core.data import InterpolationOptions
+    from gempy_engine.core.data.input_data_descriptor import InputDataDescriptor
+    from gempy_engine.core.data.interpolation_input import InterpolationInput
+    from gempy_engine.core.data.solutions import Solutions
 
 input_path = os.path.dirname(__file__) + '/../../../test/input_data'
 import sys  # These two lines are necessary only if GemPy is not installed
 
 sys.path.append("../..")
 
-
+@pytest.mark.skipif(IN_GITHUB_ACTIONS, reason="Only runs once gempy 3 is published.")
 def test_set_gempy3_gempy2_bridge():
     """
     This test is based on gempy/test/test_core/test_model.py:test_compute_model