Merge branch 'main' into sedafix

geosnap/_data.py

@@ -204,14 +204,14 @@ def seda(
             "long",
             "poolsub",
         ], "`pool` argument must be either 'pool', 'long', or 'poolsub'"
-        assert (
-            standardize
-            in [
-                "gcs",
-                "cs",
-            ]
-        ), "`standardize` argument must be either 'cs' for cohort-standardized or 'gcs' for grade-cohort-standardized"
-        fn = f"seda_{level}_{pooling}_{standardize}_4.1"
+        assert standardize in [
+            "gcs",
+            "cs",
+        ], "`standardize` argument must be either 'cs' for cohort-standardized or 'gcs' for grade-cohort-standardized"
+        if pooling=='poolsub':
+            fn = f"seda_{level}_{pooling}_{standardize}_4.1_corrected"
+        else:
+            fn = f"seda_{level}_{pooling}_{standardize}_4.1"
         local_path = pathlib.Path(self.data_dir, "seda", f"{fn}.parquet")
         remote_path = f"https://stacks.stanford.edu/file/druid:xv742vh9296/{fn}.csv"
         msg = (

geosnap/harmonize/harmonize.py

@@ -156,7 +156,7 @@ def harmonize(
     if intensive_variables is not None:
         for i in intensive_variables:
             allcols.append(i)
-   
+
     with tqdm(total=len(times), desc=f"Converting {len(times)} time periods") as pbar:
         for i in times:
             pbar.set_description(f"Harmonizing {i}")
@@ -172,10 +172,10 @@ def harmonize(
                         allocate_total=allocate_total,
                     )
                 else:
-                # if there are NaNs, tobler will raise lots of warnings, that it's filling
-                # with implicit 0s. Those warnings are superfluous most of the time
+                    # if there are NaNs, tobler will raise lots of warnings, that it's filling
+                    # with implicit 0s. Those warnings are superfluous most of the time
                     with warnings.catch_warnings():
-                        warnings.simplefilter("ignore")                 
+                        warnings.simplefilter("ignore")
                         interpolation = area_interpolate(
                             source_df,
                             target_df.copy(),
@@ -225,7 +225,8 @@ def harmonize(
             pbar.update(1)
         pbar.set_description("Complete")
         pbar.close()
-    interpolated_dfs.append(target_df[allcols].set_index(unit_index))
+    if target_year is not None:
+        interpolated_dfs.append(target_df[allcols].set_index(unit_index))
 
     harmonized_df = gpd.GeoDataFrame(pd.concat(interpolated_dfs), crs=crs)
 

geosnap/tests/test_harmonize.py

@@ -7,12 +7,6 @@
 from geosnap.harmonize import harmonize
 from geosnap.io import get_census
 
-local_raster = os.path.join(os.getcwd(), "nlcd_2011.tif")  # portability
-if not os.path.exists(local_raster):
-    p = quilt3.Package.browse("rasters/nlcd", "s3://spatial-ucr")
-    p["nlcd_2011.tif"].fetch()
-
-
 def test_harmonize_area():
     la = get_census(county_fips="06037", datastore=DataStore())
 
@@ -49,11 +43,27 @@ def test_harmonize_area_weighted():
         extensive_variables=["n_total_housing_units"],
         intensive_variables=["p_vacant_housing_units"],
         weights_method="dasymetric",
-        raster=local_raster,
+        raster='https://spatial-ucr.s3.amazonaws.com/nlcd/landcover/nlcd_landcover_2011.tif',
     )
     assert harmonized_nlcd_weighted.n_total_housing_units.sum().round(0) == 900620.0
     assert_allclose(
         harmonized_nlcd_weighted.p_vacant_housing_units.sum(),
         8832.8796,
         rtol=1e-03,
     )
+
+def test_harmonize_target_gdf():
+
+    balt = get_census(county_fips="24510", datastore=DataStore())
+    tgt_gdf = balt[balt.year==2000][['geometry']]
+    gdf = harmonize(balt,
+        target_gdf=tgt_gdf,
+        extensive_variables=["n_total_housing_units"],
+        intensive_variables=["p_vacant_housing_units"],
+    )
+    assert gdf.n_total_housing_units.sum().round(0) == 900620.0
+    assert_allclose(
+        gdf.p_vacant_housing_units.sum(),
+        8832.8796,
+        rtol=1e-03,
+    )

geosnap/visualize/mapping.py

@@ -266,6 +266,9 @@ def plot_timeseries(
             axs.format(suptitle=column)
         else:
             axs.format(suptitle=title)
+    else:
+        if title:
+            plt.suptitle(title)
 
     if save_fig:
         f.savefig(save_fig, dpi=dpi, bbox_inches="tight")