[FIX] bump matplotlib version and update usage of get_cmap (#885)

bump matplotlib version and update usage of get_cmap

nimare/reports/figures.py

@@ -211,7 +211,7 @@ def plot_coordinates(
     # Generate dictionary and array of colors for each unique ID
     ids = coordinates_df["study_id"].to_list()
     unq_ids = np.unique(ids)
-    cmap = plt.cm.get_cmap("tab20", len(unq_ids))
+    cmap = plt.colormaps["tab20"].resampled(len(unq_ids))
     colors_dict = {unq_id: mcolors.to_hex(cmap(i)) for i, unq_id in enumerate(unq_ids)}
     colors = [colors_dict[id_] for id_ in ids]
 
@@ -386,7 +386,7 @@ def plot_clusters(img, out_filename):
 
     # Define cmap depending on the number of clusters
     clust_ids = list(np.unique(img.get_fdata())[1:])
-    cmap = plt.cm.get_cmap("tab20", len(clust_ids))
+    cmap = plt.colormaps["tab20"].resampled(len(clust_ids))
 
     fig = plot_roi(
         img,

setup.cfg

@@ -42,7 +42,7 @@ install_requires =
     fuzzywuzzy  # nimare.annotate
     jinja2 # nimare.reports
     joblib>=1.3.0  # parallelization
-    matplotlib>=3.5.2  # this is for nilearn, which doesn't include it in its reqs
+    matplotlib>=3.6.0  # this is for nilearn, which doesn't include it in its reqs
     nibabel>=3.2.0  # I/O of niftis
     nilearn>=0.10.1,!=0.10.3 # https://github.com/nilearn/nilearn/pull/4256 0.10.3 is broken
     numba>=0.57.0 # used by sparse
@@ -89,7 +89,7 @@ tests =
     pytest
     pytest-cov
 minimum =
-    matplotlib==3.5.2
+    matplotlib==3.6.0
     nibabel==4.0.0
     nilearn==0.10.1
     numpy==1.22