Remove get_formatted_array

changelog.rst

@@ -20,6 +20,8 @@ User-facing changes
 
 |changed| |backwards-incompatible| The `loc::tech` and `loc::tech::carrier` sets have been removed. Model components are now indexed separately over `node`, `tech`, and `carrier` (where applicable). Although primarily an internal change, this affects the xarray dataset structure and hence how users access data in `model.inputs` and `model.results`. For example, `model.inputs.energy_cap_max.loc[{"loc_techs": "X:pv"}]` in v0.6 needs to be changed to `model.inputs.energy_cap_max.loc[{"nodes": "X", "techs": "pv"}]` in v0.7. This is functionally equivalent to first calling `model.get_formatted_array("energy_cap_max")` in v0.6, which is no longer necessary in v0.7.
 
+|changed| |backwards-incompatible| `get_formatted_array` has been removed in favour of directly accessing the data, since we no longer concatenate sets. E.g., `model.get_formatted_array("storage")` -> `model.results.storage`.
+
 |changed| |backwards-incompatible| The dimensions of the model data no longer include all possible subsets. E.g. a user can no longer access `loc_techs_supply` to view the location/technology pairs which have defined `supply` as their top-level parent. Instead, the same subset can be supplied by calling `model.inputs.inheritance.str.endswith('supply')` to create a boolean array of technologies with `supply` as their top-level parent.
 
 |changed| |backwards-incompatible| Group constraints have been removed. They will be replaced by `custom constraint` functionality.

doc/_static/notebooks/calliope_model_object.ipynb

@@ -5279,7 +5279,7 @@
    "source": [
     "# Data can also be reformatted to be easier to read (removes dimension concatenation).\n",
     "# Conversion to a pandas DataFrame is a good idea for greater readibility.\n",
-    "m.get_formatted_array(\"energy_cap\").to_pandas()"
+    "m.results.energy_cap.to_pandas()"
    ]
   },
   {
@@ -5313,9 +5313,9 @@
    "source": [
     "# >2 dimensions cannot be easily viewed in a pandas dataframe, unless a MultiIndex is used.\n",
     "# To view a 4-dimensional result, we can use `to_series()`\n",
-    "m.get_formatted_array(\n",
-    "    \"flow_out\"\n",
-    ").to_series().dropna()  # drop_na() removes all NaN values"
+    "\n",
+    "# drop_na() removes all NaN values\n",
+    "m.results.flow_out.to_series().dropna()"
    ]
   },
   {

doc/_static/notebooks/milp.ipynb

@@ -2084,7 +2084,7 @@
    ],
    "source": [
     "# We can sum operating units of CHP over all locations and turn the result into a pandas DataFrame\n",
-    "df_units = model.get_formatted_array(\"operating_units\").sum(\"nodes\").to_series()\n",
+    "df_units = model.results.operating_units.sum(\"nodes\").to_series()\n",
     "\n",
     "# The information about the dataframe tells us about the amount of data it holds in the index and in each column\n",
     "df_units.info()"

doc/_static/notebooks/national_scale.ipynb

@@ -1273,9 +1273,8 @@
     }
    ],
    "source": [
-    "# To reformat the array, deconcatenating loc_techs / loc_tech_carriers, you can use model.get_formatted_array()\n",
-    "# You can then apply loc/tech/carrier only operations, like summing information over locations:\n",
-    "model.get_formatted_array(\"resource\").sum(\"nodes\").to_series().unstack(\"techs\")"
+    "# You can apply node/tech/carrier only operations, like summing information over locations:\n",
+    "model.inputs.resource.sum(\"nodes\").to_series().unstack(\"techs\")"
    ]
   },
   {
@@ -2366,12 +2365,7 @@
    ],
    "source": [
     "# We can sum power output over all locations and turn the result into a pandas DataFrame\n",
-    "df_power = (\n",
-    "    model.get_formatted_array(\"carrier_prod\")\n",
-    "    .loc[{\"carriers\": \"power\"}]\n",
-    "    .sum(\"nodes\")\n",
-    "    .to_series()\n",
-    ")\n",
+    "df_power = model.results.flow_out.loc[{\"carriers\": \"power\"}].sum(\"nodes\").to_series()\n",
     "\n",
     "# The information about the dataframe tells us about the amount of data it holds in the index and in each column\n",
     "df_power.info()"
@@ -8310,9 +8304,7 @@
     "# notice all the NaN values which appear when seperating loc::techs to locs and techs.\n",
     "# Any NaN value means we never considered that combination of `loc` and `tech` for the variable\n",
     "\n",
-    "costs = (\n",
-    "    model.get_formatted_array(\"cost\").loc[{\"costs\": \"monetary\"}].to_series().dropna()\n",
-    ")\n",
+    "costs = model.results.cost.loc[{\"costs\": \"monetary\"}].to_series().dropna()\n",
     "costs"
    ]
   },

doc/_static/notebooks/urban_scale.ipynb

@@ -1476,9 +1476,8 @@
     }
    ],
    "source": [
-    "# To reformat the array, deconcatenating loc_techs / loc_tech_carriers, you can use model.get_formatted_array()\n",
-    "# You can then apply loc/tech/carrier only operations, like summing information over locations:\n",
-    "model.get_formatted_array(\"resource\").sum(\"locs\").to_pandas()"
+    "# You can apply node/tech/carrier only operations, like summing information over locations:\n",
+    "model.inputs.source_max.sum(\"nodes\").to_pandas()"
    ]
   },
   {
@@ -2631,13 +2630,7 @@
    ],
    "source": [
     "# We can sum heat output over all locations and turn the result into a pandas DataFrame\n",
-    "df_heat = (\n",
-    "    model.get_formatted_array(\"carrier_prod\")\n",
-    "    .loc[{\"carriers\": \"heat\"}]\n",
-    "    .sum(\"locs\")\n",
-    "    .to_pandas()\n",
-    "    .T\n",
-    ")\n",
+    "df_heat = model.results.flow_out.loc[{\"carriers\": \"heat\"}].sum(\"locs\").to_pandas().T\n",
     "\n",
     "# The information about the dataframe tells us about the amount of data it holds in the index and in each column\n",
     "df_heat.info()"
@@ -7166,7 +7159,7 @@
     "# notice all the NaN values which appear when seperating loc::techs to locs and techs.\n",
     "# Any NaN value means we never considered that combination of `loc` and `tech` for the variable\n",
     "\n",
-    "costs = model.get_formatted_array(\"cost\").loc[{\"costs\": \"monetary\"}].to_pandas()\n",
+    "costs = model.results.cost.loc[{\"costs\": \"monetary\"}].to_pandas()\n",
     "costs"
    ]
   },

src/calliope/core/model.py

@@ -510,27 +510,6 @@ def run(self, force_rerun=False, **kwargs):
         self.build(force=force_rerun)
         self.solve(force=force_rerun)
 
-    def get_formatted_array(self, var):
-        """
-        Return an xarray.DataArray with nodes, techs, and carriers as
-        separate dimensions.
-
-        Parameters
-        ----------
-        var : str
-            Decision variable for which to return a DataArray.
-
-        """
-        warnings.warn(
-            "get_formatted_array() is deprecated and will be removed in a "
-            "future version. Use `model.results[var]` instead.",
-            DeprecationWarning,
-        )
-        if var not in self._model_data.data_vars:
-            raise KeyError("Variable {} not in Model data".format(var))
-
-        return self._model_data[var]
-
     def to_netcdf(self, path):
         """
         Save complete model data (inputs and, if available, results)

src/calliope/postprocess/util.py

@@ -18,9 +18,7 @@ def subset_sum_squeeze(data, subset={}, sum_dims=None, squeeze=False):
     Parameters
     ----------
     data : xarray DataArray
-        A Calliope model data variable, either input or output, which has been
-        reformatted to deconcatenate loc_techs (or loc_tech_carriers/loc_carriers)
-        using calliope.Model().get_formatted_array(original_data)
+        A Calliope model data variable.
     subset : dict, default {}
         key:value pairs for indexing data. Uses xarray `loc[]` to index.
     sum_dims : str or list of strings, default None

tests/test_core_future_warnings.py

@@ -5,14 +5,6 @@
 
 
 class TestDeprecationWarnings:
-    def test_get_formatted_array_deprecationwarning(self):
-        model = build_model(scenario="simple_supply,one_day,investment_costs")
-
-        with pytest.warns(DeprecationWarning) as warning:
-            model.get_formatted_array("source_max")
-
-        assert check_error_or_warning(warning, "get_formatted_array() is deprecated")
-
     def test_run_deprecationwarning(self):
         model = build_model(scenario="simple_supply,two_hours,investment_costs")
 

tests/test_example_models.py

@@ -99,8 +99,8 @@ def test_nationalscale_example_results_glpk(self):
 
     def test_considers_supply_generation_only_in_total_levelised_cost(self):
         # calculation of expected value:
-        # costs = model.get_formatted_array("cost").sum(dim="locs")
-        # gen = model.get_formatted_array("flow_out").sum(dim=["timesteps", "locs"])
+        # costs = model.inputs.cost.sum(dim="locs")
+        # gen = model.inputs.flow_out.sum(dim=["timesteps", "locs"])
         # lcoe = costs.sum(dim="techs") / gen.sel(techs=["ccgt", "csp"]).sum(dim="techs")
         model = calliope.examples.national_scale()
         model.build()